{"templateName":"fundamentals-template","cssClassNames":"fundamentals-page page basicpage summit-page","allowedRenditionsWidth":["320","480","640","768","960","1200","1440","1920"],"description":"Learn how an artificial neural network works, see examples and applications, and explore the different types used in deep learning.","language":"fr","title":"What Is a Neural Network? How They Work & Why It Matters","analyticsPageType":"homepage","analyticsCategory":"general","analyticsSubCategory":"","excludeFromAnalytics":false,":mappedPath":"/fr/artificial-intelligence/machine-learning/neural-network/",":type":"snowflake-site/components/structure/page",":items":{"root":{"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"experiencefragment-banner":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-header":"aem-GridColumn aem-GridColumn--default--12","responsivegrid":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-footer":"aem-GridColumn aem-GridColumn--default--12 aem-GridColumn--offset--default--0 aem-GridColumn--default--none","experiencefragment":"aem-GridColumn aem-GridColumn--default--12 aem-GridColumn--offset--default--0 aem-GridColumn--default--none","markup_editor":"aem-GridColumn aem-GridColumn--default--12","modal_container":"aem-GridColumn aem-GridColumn--default--12"},":items":{"experiencefragment-banner":{"id":"experiencefragment-d19aed524a","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/pushdown-banner/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/pushdown-banner/master.xfmodel.json?callerPage=/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network"},"experiencefragment-header":{"id":"experiencefragment-9ee281dbc5","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/mega-nav-header/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/mega-nav-header/master.xfmodel.json?callerPage=/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network"},"markup_editor":{"id":"markup-editor-b82e12fd07","title":" ","cssContent":"div.snowflake-breadcrumb a.snowflake-breadcrumb-item,.snowflake-breadcrumb div.snowflake-breadcrumb-item{text-transform:none;font-weight:500}.snowflake-breadcrumb svg{display:none !important}.snowflake-breadcrumb a:has(svg)::after{content:'/';margin:0 12px;color:#666}.fundamentals-hero .display-2-v2{text-transform:none !important}@media screen and (min-width:1024px){.fundamentals-hero .snowflake-hero-system-inner snowflake-container snowflake-hero-system-layout-60-40{width:80% !important;max-width:700px !important}}.snowflake-hero-system-buttons-container+div:has(.snowflake-manual-breadcrumbs),.snowflake-hero-system-buttons-container+div:has(.fundamentals-hero__breadcrumbs),.snowflake-hero-system-buttons-container+div:has(.snowflake-breadcrumb){order:-1}.fundamentals-hero__breadcrumbs ol li:not(:last-child)::after{content:'';display:inline-block;width:12px;height:12px;background-image:url(\"data:image/svg+xml,%3Csvg xmlns='http://www.w3.org/2000/svg' fill='none' viewBox='0 0 10 10' data-testid='button-link-icon' class='link-icon'%3E%3Cpath d='m1.572 9.515 4.417-4.447L1.497.548' stroke='%23666'%3E%3C/path%3E%3C/svg%3E\");background-size:contain;background-repeat:no-repeat;background-position:center;margin:0 12px}.fundamentals-hero__breadcrumbs ol{margin:0;padding:0;list-style-type:none;display:flex;flex-direction:row;align-items:center}#subNav .subnav__item.subnav__item--features{color:var(--ui-01)}#subNav .subnav__item.subnav__item--features::after{content:'';display:block;width:100%;height:4px;margin:0 12px;background:var(--ui-01);position:absolute;bottom:-18.5px;left:0}.use-case-hero__architecture{background:#fff;border-radius:8px}@media screen and (min-width:768px){.use-case-body\u003E.snowflake-flexible-column-container-items\u003Ediv:first-child{position:sticky;top:200px}}.page-toc ul{list-style-type:none;padding:0}.page-toc li{padding:8px 16px;border-left:4px solid var(--ui-01);cursor:pointer;transition:300ms ease all}.page-toc li:hover{color:var(--ui-01);border-color:#7fd3f1;transition:300ms ease all}.story-highlights{padding:48px;border-radius:4px}.logo-container{max-width:180px}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container{align-items:center;justify-content:center;gap:48px;flex-wrap:nowrap}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container\u003Ediv{width:auto;margin:0 !important}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container\u003Ediv:last-child{flex-grow:1}.flex-container .aem-Grid::before,.flex-container .aem-Grid::after{display:none !important}.use-case-body table{margin-top:24px;margin-bottom:24px;width:100%;background-color:var(--ui-background-01);border-collapse:collapse;border:2px solid var(--ui-background-09);font-family:'Lato',sans-serif;color:var(--ui-background-09)}.use-case-body table thead{background-color:var(--ui-01)}.use-case-body th,.use-case-body td{min-width:120px;border:2px solid var(--ui-background-09);padding:var(--spacing-01)}.use-case-body ol{margin-top:0 !important}.use-case-body ol li{margin-bottom:1rem !important}#subNav .subnav__item.subnav__item--features{color:var(--ui-01)}#subNav .subnav__item.subnav__item--features::after{content:'';display:block;width:100%;height:4px;background:var(--ui-01);position:absolute;bottom:-18.5px;left:0}.use-case-hero__architecture{background:#fff;border-radius:8px}@media screen and (min-width:768px){.use-case-body\u003E.snowflake-flexible-column-container-items\u003Ediv:first-child{position:sticky;top:200px}}.page-toc ul{list-style-type:none;padding:0}.page-toc li{padding:8px 16px;border-left:4px solid var(--ui-01);cursor:pointer;transition:300ms ease all}.page-toc li:hover{color:var(--ui-01);border-color:#7fd3f1;transition:300ms ease all}.story-highlights{padding:48px;border-radius:4px}.logo-container{max-width:180px}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container{align-items:center;justify-content:center;gap:48px;flex-wrap:nowrap}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container\u003Ediv{width:auto;margin:0 !important}.flex-container\u003E.container\u003E.cmp-container\u003E.aem-container\u003Ediv:last-child{flex-grow:1}.flex-container .aem-Grid::before,.flex-container .aem-Grid::after{display:none !important}.snowflake-highlights{overflow:hidden;background-color:var(--ui-background-05);padding-left:24px;padding-right:24px;border:1px solid #ccc;border-radius:8px}div.use-case-body .snowflake-text h2,div.use-case-body .snowflake-text .heading-2-v2,div.use-case-body .snowflake-text h3,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-title-v2 .heading-3-v2,div.use-case-body .snowflake-text h4,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-title-v2 .heading-4-v2,div.use-case-body .snowflake-text h5,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-title-v2 .heading-5-v2,div.use-case-body .snowflake-text h6,div.use-case-body .snowflake-title-v2 .heading-6-v2,div.use-case-body .snowflake-text .heading-6-v2{text-transform:none !important}div.use-case-body .snowflake-text h2,div.use-case-body .snowflake-text .heading-2-v2,div.use-case-body .snowflake-text h3,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-text h4,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-text h5,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-text h6,div.use-case-body .snowflake-text .heading-6-v2{margin-top:1.5rem !important;line-height:1.1 !important}div.use-case-body .snowflake-text h3,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-title-v2 .heading-3-v2,div.use-case-body .snowflake-text h4,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-title-v2 .heading-4-v2,div.use-case-body .snowflake-text h5,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-title-v2 .heading-5-v2,div.use-case-body .snowflake-text h6,div.use-case-body .snowflake-text .heading-6-v2,div.use-case-body .snowflake-text .heading-6-v2,div.use-case-body .snowflake-title-v2 .heading-6-v2{font-family:Lato,sans-serif !important;font-weight:800 !important}div.use-case-body .snowflake-text h2,div.use-case-body .snowflake-text .heading-2-v2,div.use-case-body .snowflake-title-v2 .heading-2-v2{text-transform:none !important;font-size:28px !important}div.use-case-body .snowflake-text h3,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-title-v2 .heading-3-v2{font-size:22px !important}div.use-case-body .snowflake-text h4,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-title-v2 .heading-4-v2{font-size:18px !important}div.use-case-body .snowflake-text h5,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-title-v2 .heading-5-v2{font-size:16px !important}div.use-case-body .snowflake-text h6,div.use-case-body .snowflake-text .heading-6-v2,div.use-case-body .snowflake-title-v2 .heading-6-v2{font-size:14px !important}@media screen and (min-width:992px){div.use-case-body .snowflake-text h2,div.use-case-body .snowflake-text .heading-2-v2,div.use-case-body .snowflake-title-v2 .heading-2-v2{font-size:38px !important}div.use-case-body .snowflake-text h3,div.use-case-body .snowflake-text .heading-3-v2,div.use-case-body .snowflake-title-v2 .heading-3-v2{font-size:26px !important}div.use-case-body .snowflake-text h4,div.use-case-body .snowflake-text .heading-4-v2,div.use-case-body .snowflake-title-v2 .heading-4-v2{font-size:22px !important}div.use-case-body .snowflake-text h5,div.use-case-body .snowflake-text .heading-5-v2,div.use-case-body .snowflake-title-v2 .heading-5-v2{font-size:18px !important}div.use-case-body .snowflake-text h6,div.use-case-body .snowflake-text .heading-6-v2,div.use-case-body .snowflake-title-v2 .heading-6-v2{font-size:16px !important}}","jsContent":"window.addEventListener('click',(e)=\u003E{if(e.target.tagName==='LI'&&e.target.dataset.anchor){const target=document.getElementById(e.target.dataset.anchor);if(target){const targetPosition=target.getBoundingClientRect().top+window.pageYOffset-220;window.scrollTo({top:targetPosition,behavior:'smooth'});}}});",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false},"responsivegrid":{"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"container":"aem-GridColumn aem-GridColumn--default--12","hero_system":"aem-GridColumn aem-GridColumn--default--12"},":items":{"hero_system":{"id":"hero-system-86db2f23a4","heroStyle":"tertiary","headline":{"id":"headline","type":"display2","lines":["What Is a Neural Network? A Complete Guide"],":type":"snowflake-site/components/title-v2"},"subheadline":{"id":"subheadline","text":"\u003Cp\u003EWhat is a neural network? Learn how an artificial neural network works, see examples and applications, and explore the different types used in deep learning.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"position":"center",":type":"snowflake-site/components/hero-system","appliedCssClassNames":"snowflake-hero-system-background-grad-white"},"container":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"container":"aem-GridColumn aem-GridColumn--default--12"},"id":"container-452f7d1816",":type":"snowflake-site/components/container",":items":{"container":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"flexible_column_cont":"aem-GridColumn aem-GridColumn--default--12","related_content":"aem-GridColumn aem-GridColumn--default--12"},"id":"container-3e5cc515c4",":type":"snowflake-site/components/container",":items":{"flexible_column_cont":{"id":"flexible-column-container-e5a13aa72b","type":"2-column-25-75","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"medium","bottomPadding":"medium","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"use-case-body","backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-738ad6073e",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"text":{"id":"text-f9609caf9c","additionalClasses":"page-toc","text":"\u003Cul\u003E\r\n\u003Cli data-anchor=\"overview\"\u003EOverview\u003C/li\u003E\r\n\u003Cli data-anchor=\"a\"\u003EWhat is a neural network?\u003C/li\u003E\r\n\u003Cli data-anchor=\"use\"\u003EWhy are neural networks important?\u003C/li\u003E\r\n\u003Cli data-anchor=\"benefits\"\u003ENeural network applications and use cases\u003C/li\u003E\r\n\u003Cli data-anchor=\"how\"\u003EHow do neural networks work?\u003C/li\u003E\r\n\u003Cli data-anchor=\"ai\"\u003ETypes of neural networks\u003C/li\u003E\r\n\u003Cli data-anchor=\"tools\"\u003EExamples of neural networks in action\u003C/li\u003E\r\n\u003Cli data-anchor=\"last\"\u003EConclusion\u003C/li\u003E\r\n\u003Cli data-anchor=\"faq\"\u003ENeural network FAQs\u003C/li\u003E\r\n\u003Cli data-anchor=\"customer\"\u003ECustomers Using Snowflake\u003C/li\u003E\r\n\u003Cli data-anchor=\"resources\"\u003EResources\u003C/li\u003E\r\n\u003C/ul\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-size-regular text-color-text-05"},"experiencefragment":{"id":"experiencefragment-c58532ee16","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/share-icons/share-icons-no-title/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/share-icons/share-icons-no-title.xfmodel.json?callerPage=/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network","appliedCssClassNames":"snowflake-responsive-component-top-padding-extra-small"}},":itemsOrder":["text","experiencefragment"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"},"flexible_column_content_container_2":{"layout":"SIMPLE","id":"fundamentals-main-content",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"container_copy":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"overview",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-4f261b03cf","additionalClasses":"headline-decoration","type":"heading2","lines":["Overview"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-b9b5651647","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003ENeural networks are the fundamental technology powering today's \u003Ca href=\"https://www.snowflake.com/en/product/snowflake-intelligence/\"\u003EAI breakthroughs\u003C/a\u003E. Inspired by how neurons connect in the human brain, these systems consist of interconnected layers of artificial "neurons" (mathematical operations) that learn by analyzing massive data sets, automatically discovering patterns without being explicitly told what to look for. \u003C/p\u003E\r\n\u003Cp\u003ETheir ability to generalize from examples allows neural networks to tackle problems that were previously unsolvable with traditional computing approaches, such as the image recognition algorithms that allow self-driving cars to identify and react to road conditions in real-time, or \u003Ca href=\"https://www.snowflake.com/en/fundamentals/natural-language-processing/\"\u003Enatural language processing (NLP)\u003C/a\u003E that enables nuanced translations from one language to another.\u003C/p\u003E\r\n\u003Cp\u003EThis guide will explain how neural networks operate, break down the different types of neural networks and demonstrate why they’re a foundational technology for applications such as facial recognition and voice-driven digital assistants. \u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy_":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"container_115652453":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__480869401":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__373061683":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__1444615495":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__222865509":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__1413360324":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy_":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__455068363":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__1500909857":"aem-GridColumn aem-GridColumn--default--12","container_1858830768":"aem-GridColumn aem-GridColumn--default--12","container_copy_copy__1978652079":"aem-GridColumn aem-GridColumn--default--12"},"id":"rel",":type":"snowflake-site/components/container",":items":{"container_copy_copy_":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"a",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-3d85b33ce9","additionalClasses":"headline-decoration","type":"heading2","lines":["What is a neural network?"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-721c3ef2b8","text":"\u003Cp\u003EAn artificial neural network (ANN) is a \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/models/\"\u003Emachine learning model\u003C/a\u003E composed of interconnected processing units called neurons or nodes, organized in layers. These networks learn by example, processing large training data sets to automatically recognize patterns in the data. Through repeated exposure to examples, they adjust the connections between each set of neurons to improve accuracy, enabling them to identify complex patterns and make predictions without being explicitly programmed.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__1444615495":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"use",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-405dd8a5e5","additionalClasses":"headline-decoration","type":"heading2","lines":["Why are neural networks important?"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-1badebbb92","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003EUnlike conventional software that requires explicit rules, neural networks excel at pattern recognition by learning directly from examples. That allows them to solve complex problems involving unstructured data — such as images, audio and text — that are extremely difficult or impossible for traditional programming to handle. This pattern recognition capability is the foundation for critical real-world tasks: identifying objects in images, understanding human speech and detecting subtle anomalies in massive data sets. Their ability to find hidden patterns in messy, unstructured data makes them indispensable for problems where the rules are too complex to code manually.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__1978652079":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"benefits",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-6ca668b17f","additionalClasses":"headline-decoration","type":"heading2","lines":["Neural network applications and use cases"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-788b67117f","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003ENeural networks have been deployed across a wide range of domains. Here are six fields where ANNs have had a significant real-world impact:\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EComputer vision\u003C/h3\u003E\r\n\u003Cp\u003ENeural networks enable \u003Ca href=\"https://www.snowflake.com/en/fundamentals/machine-learning-frameworks/\"\u003Emachines\u003C/a\u003E to interpret and understand visual information from images and videos. Popular applications include facial recognition, medical image analysis, autonomous vehicle navigation and quality control in manufacturing.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ENatural language processing\u003C/h3\u003E\r\n\u003Cp\u003EThese systems process and understand \u003Ca href=\"https://www.snowflake.com/en/fundamentals/ai-in-advertising/\"\u003Ehuman language\u003C/a\u003E, powering machine translation, chatbots, sentiment analysis and text generation. NLP systems powered by ANNs have revolutionized how we interact with technology through voice assistants and automated customer service bots.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ERecommendation engines \u003C/h3\u003E\r\n\u003Cp\u003ENeural networks analyze user behavior and preferences to suggest personalized content, products or services. Platforms like Netflix, Amazon and Spotify use these systems to drive engagement and sales.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EAnomaly detection systems\u003C/h3\u003E\r\n\u003Cp\u003EThese networks identify unusual patterns that deviate from normal behavior in data streams. They're critical for detecting fraudulent financial transactions, identifying potential \u003Ca href=\"https://www.snowflake.com/en/customers/all-customers/case-study/merkle/\"\u003Ecybersecurity threats\u003C/a\u003E and predicting equipment failures in industrial settings.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EHealthcare and drug discovery\u003C/h3\u003E\r\n\u003Cp\u003ENeural networks help medical professionals diagnose diseases, create treatment plans and analyze medical imaging with accuracy rivaling human experts. They also accelerate drug discovery by predicting molecular interactions and identifying promising compounds.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ESpeech recognition and synthesis\u003C/h3\u003E\r\n\u003Cp\u003EThese systems convert spoken language into text and generate natural-sounding speech from text. ANNs power virtual assistants, transcription services and accessibility tools for individuals with disabilities.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__455068363":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"how",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-0be6e5dc97","additionalClasses":"headline-decoration","type":"heading2","lines":["How do neural networks work?"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-d3d89377bd","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003EAll neural networks are composed of the same fundamental elements. They include:\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EThe layers\u003C/h3\u003E\r\n\u003Cp\u003ENeural networks are organized into three types of layers: an input layer that receives the raw data, one or more hidden layers that process the information, and an output layer that produces the final result. Information flows forward through the network, with each layer \u003Ca href=\"https://www.snowflake.com/en/fundamentals/unlocking-value-through-data-transformation-in-modern-pipelines/\"\u003Etransforming the data\u003C/a\u003E and passing it to the next layer. The hidden layers are where the network learns to recognize increasingly complex patterns. For example, early layers might detect simple features like edges in an image, while deeper layers identify complex objects like faces or cars.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ENeurons, weights and biases\u003C/h3\u003E\r\n\u003Cp\u003ENeurons are the basic processing units that receive multiple inputs, perform a calculation and pass the result forward to the next layer. Weights determine how important each input is to a neuron's calculation — think of them as volume controls that amplify or diminish each signal. Biases help adjust the neuron's sensitivity, acting as a baseline that allows the network to fit complex patterns in the data by making neurons more or less likely to activate.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EThe training process\u003C/h3\u003E\r\n\u003Cp\u003ETraining involves showing the network many labeled examples and letting it make predictions, then measuring how far those predictions deviate from the correct answers. The network uses these errors to adjust its weights and biases slightly in the direction that improves accuracy, tracing backward through the layers to determine what changes will help most. This process repeats thousands or millions of times across the entire data set until the network learns to recognize patterns and make accurate predictions on new data it hasn't seen before.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__222865509":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","text":"aem-GridColumn aem-GridColumn--default--12"},"id":"ai",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-f94beddc8a","additionalClasses":"headline-decoration","type":"heading2","lines":["Types of neural networks"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text":{"id":"text-9b7970161f","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003EThere are half a dozen different kinds of ANNs, each designed to excel at specific tasks. Here are the most widely used:\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EFeedforward neural networks\u003C/h3\u003E\r\n\u003Cp\u003EWith feedforward neural networks (FNNs), information flows in a single direction, from input to output, without looping back. These networks are used for basic classification and regression tasks where the sequence of the input data is not important. In other words, FNNs are useful for tasks like predicting house prices, classifying emails as spam or recognizing simple patterns in tabular data, but they would not be used for speech recognition or image classification.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EConvolutional neural networks\u003C/h3\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/neural-network/convolutional-neural-network/\"\u003EConvolutional neural networks\u003C/a\u003E (CNNs) are specifically designed to process grid-like data such as images, using specialized layers that scan across the input to detect local patterns like edges, textures and shapes. They're highly efficient because they learn to recognize the same features anywhere in an image, rather than treating each pixel position as completely independent. CNNs power most modern computer vision applications, from facial recognition and medical image analysis to the environmental perception systems in self-driving cars.\u003C/p\u003E\r\n\u003Ch3\u003EGraph neural networks\u003C/h3\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/neural-network/graph-neural-network/\"\u003EGraph neural networks\u003C/a\u003E (GNNs) are designed to process data structured as a network of connected entities, where nodes represent things like users, accounts, products or molecules, and edges represent the relationships between them. Rather than treating each data point as independent, GNNs learn from both an entity’s own features and the surrounding network of connections. This makes them useful for tasks where relationships carry important predictive signal, such as fraud detection, recommendation systems, drug discovery, entity resolution and supply chain analysis.\u003C/p\u003E\r\n\u003Ch3\u003ERecurrent neural networks\u003C/h3\u003E\r\n\u003Cp\u003EUnlike FNNs, recurrent neural networks (RNNs) are built to handle sequential data where order matters, such as text, speech or time-series data. Their ability to remember previous inputs allows RNNs to use context from earlier in the sequence to inform current predictions. They're used in applications like language translation, speech recognition and predicting stock prices based on historical trends.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EGenerative adversarial networks\u003C/h3\u003E\r\n\u003Cp\u003EGeneral adversarial networks (GANs) consist of two neural networks that compete against each other: One network generates fake data (like images or audio) while the other tries to discriminate real data from fake. Through this competition, the generator becomes increasingly skilled at creating realistic outputs that can fool the discriminator. GANs are used to create synthetic images, generate realistic voices, enhance photo resolution and even create deepfakes.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ETransformer networks\u003C/h3\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/neural-network/transformer/\"\u003ETransformer\u003C/a\u003E networks use an attention mechanism that allows them to weigh the importance of different parts of the input when making predictions rather than processing information sequentially. This architecture excels at understanding context and relationships in language, making it ideal for tasks where long-range dependencies matter. Transformers power most \u003Ca href=\"https://www.snowflake.com/en/fundamentals/ai-programming-languages/\"\u003Emodern language models\u003C/a\u003E, including chatbots, translation systems and text generation tools like GPT.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EAutoencoders\u003C/h3\u003E\r\n\u003Cp\u003E\u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/machine-learning/neural-network/autoencoder/\"\u003EAutoencoders\u003C/a\u003E are networks designed to compress data into a compact representation and then reconstruct it back to its original form, learning the most important features in the process. They're trained to recreate their input as accurately as possible, which forces them to capture the essential patterns while filtering out noise. These networks are used for data compression, removing noise from images, detecting anomalies and generating new variations of existing data.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","text"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__1413360324":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2_copy":"aem-GridColumn aem-GridColumn--default--12","text_copy":"aem-GridColumn aem-GridColumn--default--12"},"id":"tools",":type":"snowflake-site/components/container",":items":{"title_v2_copy":{"id":"title-v2-d98943da07","additionalClasses":"headline-decoration","type":"heading2","lines":["Examples of neural networks in action"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text_copy":{"id":"text-19f3166012","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003EIt’s increasingly difficult to find digital tools that don’t have some connection to neural networks. Here are some common everyday applications made possible by this technology:\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EFacial recognition\u003C/h3\u003E\r\n\u003Cp\u003EYour smartphone relies on neural networks to identify your face and unlock the device, analyzing facial features and comparing them to stored data. Social media platforms employ similar technology to automatically tag people in photos by recognizing their faces. Security systems and airports also use facial recognition for identity verification and \u003Ca href=\"https://www.snowflake.com/en/fundamentals/rbac/\"\u003Eaccess contro\u003C/a\u003El.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EVoice assistants\u003C/h3\u003E\r\n\u003Cp\u003EDigital assistants like Siri, Alexa and Google Assistant rely on neural networks to convert your spoken words into text and understand the context of what you're saying. These systems process the audio patterns of your voice, interpret your intent and generate appropriate responses. They continuously improve by learning from millions of voice interactions across different accents and speaking styles.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EEmail spam filters\u003C/h3\u003E\r\n\u003Cp\u003ENeural networks analyze the content, sender information and patterns in emails to determine whether messages are legitimate or spam. They learn to recognize common spam characteristics like suspicious links, deceptive subject lines and typical phishing language. These filters adapt over time as spammers change their tactics, protecting your inbox from unwanted and malicious messages.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EStreaming service recommendations\u003C/h3\u003E\r\n\u003Cp\u003ENetflix, Spotify and YouTube use neural networks to analyze your viewing or listening history and suggest content you might enjoy. These systems identify patterns in the media you consume, compare your preferences with similar users and predict what will keep you engaged. The recommendations become more personalized as the system learns more about your tastes over time.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ENavigation and traffic prediction\u003C/h3\u003E\r\n\u003Cp\u003EMapping apps like Google Maps and Waze use neural networks to predict traffic conditions and suggest the fastest route to your destination. These systems analyze real-time data from millions of users, historical traffic patterns and current road conditions to forecast delays. They continuously update predictions as conditions change, helping you avoid congestion and arrive on time.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003ESocial media content moderation\u003C/h3\u003E\r\n\u003Cp\u003EPlatforms like Facebook, Instagram and YouTube use neural networks to automatically detect and remove harmful content such as hate speech, violent images and misinformation. These systems scan millions of posts, images and videos every day, flagging content that violates community guidelines for human review. While far from perfect, these moderation tools help keep platforms safer by catching a large amount of problematic content before it spreads widely.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Ch3\u003EAutocorrect and predictive text\u003C/h3\u003E\r\n\u003Cp\u003EYour smartphone's keyboard uses neural networks to correct spelling mistakes and predict the next word you're likely to type. These systems learn from your typing patterns and common language usage to offer relevant suggestions. They adapt to your personal writing style, including frequently used words and phrases unique to you.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2_copy","text_copy"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_1858830768":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{},"id":"solutions",":type":"snowflake-site/components/container",":items":{},":itemsOrder":[],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"},"container_115652453":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"text_copy_copy_copy_":"aem-GridColumn aem-GridColumn--default--12","title_v2_copy_copy_c":"aem-GridColumn aem-GridColumn--default--12"},"id":"last",":type":"snowflake-site/components/container",":items":{"title_v2_copy_copy_c":{"id":"title-v2-59838da1f1","additionalClasses":"headline-decoration","type":"heading2","lines":["Conclusion"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"text_copy_copy_copy_":{"id":"text-ee99dee088","additionalClasses":"list--blue-bullets","text":"\u003Cp\u003EArtificial neural networks are the foundational technology behind \u003Ca href=\"https://www.snowflake.com/en/artificial-intelligence/\"\u003Emodern AI\u003C/a\u003E, enabling machines to learn from data and perform \u003Ca href=\"https://www.snowflake.com/en/fundamentals/generative-ai/\"\u003Ecomplex tasks\u003C/a\u003E once thought exclusive to humans. Modeled after the human brain, these networks excel at recognizing patterns in unstructured data like images, speech and text without explicit programming. Their impact can be felt almost everywhere, from facial recognition and voice assistants to recommendation systems and spam filters, all using different architectures designed to address specific problems. \u003C/p\u003E\r\n\u003Cp\u003EWhat makes ANNs powerful is their ability to automatically discover patterns in massive data sets by adjusting millions of parameters through iterative learning. As computational power and data availability grow, neural networks will continue to expand their capabilities and shape the future of technology and society.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text","appliedCssClassNames":"text-color-text-05 snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2_copy_copy_c","text_copy_copy_copy_"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"},"container_copy_copy__1500909857":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"title_v2":"aem-GridColumn aem-GridColumn--default--12","simple_snowflake_acc":"aem-GridColumn aem-GridColumn--default--12"},"id":"faq",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-f262145226","additionalClasses":"headline-decoration","type":"heading2","lines":["Neural network FAQs"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"simple_snowflake_acc":{"id":"simple-snowflake-accordion-426f5913ac","additionalClasses":"list--blue-bullets","showDivider":false,"accordionItemsList":[{"title":"How are neural networks different from regular computer programs?","richText":"\u003Cp\u003ETraditional programs follow explicit rules written by programmers for every situation, while neural networks learn patterns from examples and figure out the rules themselves. This makes neural networks better at handling complex, messy problems like recognizing faces or understanding speech, where writing all the rules manually would be impossible.\u003C/p\u003E\r\n"},{"title":"Do neural networks actually think like human brains?","richText":"\u003Cp\u003ENo, neural networks are only loosely inspired by biological brains and work very differently in practice. While both use interconnected units to process information, neural networks are mathematical models running on computers, not biological neurons, and they lack consciousness, emotions or true understanding.\u003C/p\u003E\r\n"},{"title":"How much data do neural networks need to learn?","richText":"\u003Cp\u003EThe amount of data varies widely depending on the task's complexity — simple problems might need thousands of examples, while complex tasks like language understanding can require millions or billions. The general rule is that more complex patterns require more data, though techniques like transfer learning allow networks to apply knowledge from one task to another, reducing data requirements.\u003C/p\u003E\r\n"}],":type":"snowflake-site/components/simple-snowflake-accordion","appliedCssClassNames":"snowflake-responsive-component-bottom-padding-small"}},":itemsOrder":["title_v2","simple_snowflake_acc"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__480869401":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"flexible_column_cont":"aem-GridColumn aem-GridColumn--default--12","title_v2":"aem-GridColumn aem-GridColumn--default--12"},"id":"customer",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-7469381a11","additionalClasses":"headline-decoration","type":"heading2","lines":["Customers using Snowflake"],":type":"snowflake-site/components/title-v2","appliedCssClassNames":"snowflake-responsive-component-top-padding-none"},"flexible_column_cont":{"id":"flexible-column-container-91688058f1","type":"2-column-even","alignColumns":"match-height","containerMaxWidth":"extra-large","topPadding":"extra-small","bottomPadding":"small","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-d1ea5a4ed7",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"card_v2":{"id":"card-v2-d2ab3b13d2","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","title":{"id":"title","type":"heading4","lines":["Simon Data Evolves Marketing with Composable AI Agents Built on Snowflake Cortex AI"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"attributes":{"target":"_blank"},"url":"https://www.snowflake.com/en/customers/all-customers/case-study/simon-data/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read the story"},"image":{"id":"image","height":"720","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--8d3027e6-8989-479b-a8da-43423088b3ed/simon-data-customer-card%25402x.jpg?preferwebp=true&quality=85","alt":"simon data logo","lazyEnabled":true,"width":"1680",":type":"snowflake-site/components/image"},"type":"content-card","text":{"id":"text","text":"\u003Cp\u003EWith Snowflake as its foundation for agentic AI, Simon Data helps marketers boost revenue by delivering contextual personalization at scale — all without moving data or compromising governance.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"vertical"}},":itemsOrder":["card_v2"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"},"flexible_column_content_container_2":{"layout":"SIMPLE","id":"container-c5b4fde481",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"card_v2_copy":{"id":"card-v2-d2bc85befa","configurationStatus":{"configured":true,"message":""},":type":"snowflake-site/components/card-v2","title":{"id":"title","type":"heading4","lines":["WHOOP Improves AI/ML Financial Forecasting While Enhancing Members’ Experiences"],":type":"snowflake-site/components/title-v2"},"button":{"id":"button","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/customers/all-customers/case-study/whoop/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Read the story"},"image":{"id":"image","height":"720","src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--15863a57-5aeb-4400-b348-edd4aedb6520/whoop-customer-card%25402x.jpg?preferwebp=true&quality=85","alt":"penske logo","lazyEnabled":true,"width":"1680",":type":"snowflake-site/components/image"},"type":"content-card","text":{"id":"text","text":"\u003Cp\u003EWith Snowflake and Apache Iceberg, WHOOP teams have centralized access to data while reducing complexity, lowering costs and improving critical processes.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"},"layoutStyle":"vertical"}},":itemsOrder":["card_v2_copy"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"},":type":"snowflake-site/components/flexible-column-container","isBlogPage":false,"isActiveTOC":false}},":itemsOrder":["title_v2","flexible_column_cont"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"container_copy_copy__373061683":{"layout":"RESPONSIVE_GRID","columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","columnClassNames":{"flexible_column_cont":"aem-GridColumn aem-GridColumn--default--12","title_v2":"aem-GridColumn aem-GridColumn--default--12"},"id":"resources",":type":"snowflake-site/components/container",":items":{"title_v2":{"id":"title-v2-e256db7c0f","type":"heading2","lines":["AI and ML Resources"],":type":"snowflake-site/components/title-v2"},"flexible_column_cont":{"id":"flexible-column-container-1edd0b2736","type":"2-column-even","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"extra-small","bottomPadding":"none","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-bb5dcc32a1",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"content_chip_copy_co":{"id":"content-chip-d9f00d8f3b","tagText":"report","tagColor":"#C6EDF1","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/resource/data-strategies-for-ai-leaders/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Get the report"},"headline":{"id":"title","type":"heading5","lines":["Data Strategies for AI Leaders"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"},"content_chip_copy_co_1958646508":{"id":"content-chip-d90f632f5e","tagText":"Ebook","tagColor":"#71D3DC","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"attributes":{"target":"_blank"},"url":"https://www.snowflake.com/resource/a-practical-guide-to-ai-agents"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Get the ebook"},"headline":{"id":"title","type":"heading5","lines":["A Practical Guide to AI Agents "],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"},"content_chip_copy_co_2069341087":{"id":"content-chip-d3473631d7","tagText":"Ebook","tagColor":"#71D3DC","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"attributes":{"target":"_blank"},"url":"https://www.snowflake.com/resource/5-ways-ai-and-machine-learning-accelerate-b2b-marketing-roi/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Get the ebook"},"headline":{"id":"title","type":"heading5","lines":["5 Ways AI and Machine Learning Accelerate B2B Marketing ROI"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"}},":itemsOrder":["content_chip_copy_co","content_chip_copy_co_1958646508","content_chip_copy_co_2069341087"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"},"flexible_column_content_container_2":{"layout":"SIMPLE","id":"container-d3887ae646",":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container",":items":{"content_chip_copy_co_360891082":{"id":"content-chip-84e97c12cf","tagText":"Ebook","tagColor":"#71D3DC","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"attributes":{"target":"_blank"},"url":"https://www.snowflake.com/en/lp/radical-roi-generative-ai/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Get the ebook"},"headline":{"id":"title","type":"heading5","lines":["The Radical ROI of Gen AI"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"},"content_chip_copy_co":{"id":"content-chip-edd9c4ecb0","tagText":"Academy","tagColor":"#99AEB5","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"attributes":{"target":"_blank"},"url":"https://www.snowflake.com/data-cloud-academy-generative-ai-llm/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Enroll now"},"headline":{"id":"title","type":"heading5","lines":["Generative AI & ML School"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"},"content_chip_copy_co_1390138041":{"id":"content-chip-3aeec367dc","tagText":"report","tagColor":"#C6EDF1","cta":{"id":"cta","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"https://www.snowflake.com/en/lp/snowflake-ai-data-predictions/"},"linkTargetContentType":"GENERIC",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_EXTERNAL","text":"Get the report"},"headline":{"id":"title","type":"heading5","lines":["Snowflake AI + Data Predictions 2026"],":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/content-chip"}},":itemsOrder":["content_chip_copy_co_360891082","content_chip_copy_co","content_chip_copy_co_1390138041"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-medium"},":type":"snowflake-site/components/flexible-column-container","isBlogPage":false,"isActiveTOC":false}},":itemsOrder":["title_v2","flexible_column_cont"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"}},":itemsOrder":["container_copy_copy_","container_copy_copy__1444615495","container_copy_copy__1978652079","container_copy_copy__455068363","container_copy_copy__222865509","container_copy_copy__1413360324","container_1858830768","container_115652453","container_copy_copy__1500909857","container_copy_copy__480869401","container_copy_copy__373061683"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-extra-small"},"markup_editor":{"id":"markup-editor-d2003eed40","title":"Page CSS","cssContent":".use-case-body table{width:100%;background-color:var(--ui-background-01);border-collapse:collapse;border:2px solid var(--ui-background-09);font-family:'Lato',sans-serif;color:var(--ui-background-09)}.use-case-body table thead{background-color:var(--ui-01)}.use-case-body th,.use-case-body td{border:2px solid var(--ui-background-09);padding:var(--spacing-01)}.use-case-body ol{margin-top:0 !important}.use-case-body ol li{margin-bottom:1rem !important}#subNav .subnav__item.subnav__item--features{color:var(--ui-01)}#subNav .subnav__item.subnav__item--features::after{content:'';display:block;width:100%;height:4px;background:var(--ui-01);position:absolute;bottom:-18.5px;left:0}.use-case-hero__architecture{background:#fff;border-radius:8px}@media screen and (min-width:768px){.use-case-body \u003E .snowflake-flexible-column-container-items \u003E div:first-child{position:sticky;top:200px}}.page-toc ul{list-style-type:none;padding:0}.page-toc li{padding:8px 16px;border-left:4px solid var(--ui-01);cursor:pointer;transition:300ms ease all}.page-toc li:hover{color:var(--ui-01);border-color:#7fd3f1;transition:300ms ease all}.story-highlights{padding:48px;border-radius:4px}.logo-container{max-width:180px}.flex-container \u003E .container \u003E .cmp-container \u003E .aem-container{align-items:center;justify-content:center;gap:48px;flex-wrap:nowrap}.flex-container \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div{width:auto;margin:0 !important}.flex-container \u003E .container \u003E .cmp-container \u003E .aem-container \u003E div:last-child{flex-grow:1}.flex-container .aem-Grid::before,.flex-container .aem-Grid::after{display:none !important}.snowflake-highlights{overflow:hidden;background-color:var(--ui-background-05);padding-left:24px;padding-right:24px;border:1px solid #ccc;border-radius:8px}","jsContent":"window.addEventListener('click',(e)=\u003E{if(e.target.tagName==='LI'&&e.target.dataset.anchor){const target=document.getElementById(e.target.dataset.anchor);if(target){const targetPosition=target.getBoundingClientRect().top+window.pageYOffset-220;window.scrollTo({top:targetPosition,behavior:'smooth'});}}});",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false}},":itemsOrder":["container_copy","container_copy_copy_","markup_editor"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-large"},":type":"snowflake-site/components/flexible-column-container","isBlogPage":false,"isActiveTOC":false},"related_content":{"id":"related-content-3472a1ccdd","relatedContent":[],":type":"snowflake-site/components/blog/related-content","isBlogPage":false,"appliedCssClassNames":"snowflake-responsive-component-bottom-padding-medium"}},":itemsOrder":["flexible_column_cont","related_content"],"appliedCssClassNames":"snowflake-container snowflake-responsive-container-inner-padding-small"}},":itemsOrder":["container"],"appliedCssClassNames":"snowflake-responsive-container-inner-padding-small"}},":itemsOrder":["hero_system","container"],":type":"wcm/foundation/components/responsivegrid"},"modal_container":{"layout":"SIMPLE","id":"container-427d8fbe17",":type":"snowflake-site/components/modal/modal-container",":items":{},":itemsOrder":[]},"experiencefragment-footer":{"id":"experiencefragment-5f23da12d4","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/footer/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/fr/site/footer/master.xfmodel.json?callerPage=/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network"},"experiencefragment":{"id":"experiencefragment-16bfd85f37","configured":false,":type":"snowflake-site/components/experiencefragment"}},":itemsOrder":["experiencefragment-banner","experiencefragment-header","markup_editor","responsivegrid","modal_container","experiencefragment-footer","experiencefragment"],":type":"wcm/foundation/components/responsivegrid"}},":itemsOrder":["root"],":hierarchyType":"page",":path":"/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network","isPasswordProtected":false,"analyticsContentTags":["snowflake-site:taxonomy/content-type/fundamentals","snowflake-site:taxonomy/product/ai"],"analyticsEnabled":true,"coveoConfig":{"pipeline":"snowflake.com","apiKey":"xx335921a6-2a0a-40f2-a167-e390b4766c3d","organizationId":"snowflakecomputingproduction8neljofn","searchHub":"snowflake.com"},"analyticsDebugMode":false,"analyticsData":{"excludeFromAnalytics":false,"subCategory":"","pageType":"homepage","templateName":"fundamentals-template","siteName":"snowflake","pageUrl":"/content/snowflake-site/global/fr/artificial-intelligence/machine-learning/neural-network","language":"fr","category":"general","pageName":"What Is a Neural Network? A Complete Guide","contentTags":["snowflake-site:taxonomy/content-type/fundamentals","snowflake-site:taxonomy/product/ai"]},"locale":"fr"}