{"id":847225,"date":"2022-05-24T09:11:51","date_gmt":"2022-05-24T16:11:51","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-blog-post&#038;p=847225"},"modified":"2022-08-16T13:48:33","modified_gmt":"2022-08-16T20:48:33","slug":"empower-ai-developers-2","status":"publish","type":"msr-blog-post","link":"https:\/\/www.microsoft.com\/en-us\/research\/articles\/empower-ai-developers-2\/","title":{"rendered":"Empower AI developers"},"content":{"rendered":"\n<div class=\"wp-block-media-text has-vertical-padding-none  alignwide has-media-on-the-right is-stacked-on-mobile is-vertically-aligned-top is-style-spectrum is-style-border is-style-offset-media--top\" data-bi-an=\"media-text\"><figure class=\"wp-block-media-text__media\"><img loading=\"lazy\" decoding=\"async\" width=\"800\" height=\"450\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg.png\" alt=\"GPU energy usage graph - showing 4 out of 8 GPUs\" class=\"wp-image-840610 size-full\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg.png 800w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-300x169.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-768x432.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-655x368.png 655w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-343x193.png 343w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-240x135.png 240w, https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2022\/04\/Reducing-AI-carbon-footprint_GPU-graph_1400x788.jpg-640x360.png 640w\" sizes=\"(max-width: 800px) 100vw, 800px\" \/><\/figure><div class=\"wp-block-media-text__content\" data-bi-an=\"media-text\">\n<p>Progress in machine learning is measured in part through the constant improvement of performance metrics such as accuracy or latency. Carbon footprint metrics, while being an equally important target, have not received the same degree of attention. With contributions from our research team, <strong>Azure ML<\/strong> now provides transparency around machine learning resource utilization, including GPU energy consumption and computational cost, for both training and inference at scale. This reporting can raise developers\u2019 awareness of the carbon cost of their model development process and encourage them to optimize their experimentation strategies.<\/p>\n\n\n\n<p><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/techcommunity.microsoft.com\/t5\/green-tech-blog\/charting-the-path-towards-sustainable-ai-with-azure-machine\/ba-p\/2866923\" target=\"_blank\" rel=\"noreferrer noopener\">Read the blog ><span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/p>\n<\/div><\/div>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1400\" height=\"788\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/10\/1400x788_Archai_NoLogo-2.gif\" alt=\"An animated illustration of the neural architecture search platform Archai automatically identifying neural network architectures for a given dataset.\" class=\"wp-image-695670\"\/><figcaption>An animated illustration of the neural architecture search platform Archai automatically identifying neural network architectures for a given dataset.<\/figcaption><\/figure>\n\n\n\n<div class=\"annotations \" data-bi-aN=\"margin-callout\">\n\t<ul class=\"annotations__list card depth-16 bg-body p-4 annotations__list--right\">\n\t\t<li class=\"annotations__list-item\">\n\t\t\t\t\t\t<span class=\"annotations__type d-block text-uppercase font-weight-semibold text-neutral-300 small\">Blog<\/span>\n\t\t\t<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/blog\/archai-can-design-your-neural-network-with-state-of-the-art-neural-architecture-search-nas\/\" target=\"_self\" class=\"annotations__link font-weight-semibold text-decoration-none\" data-bi-type=\"annotated-link\" aria-label=\"Archai can design your neural network with state-of-the-art neural architecture search (NAS)\" data-bi-aN=\"margin-callout\" data-bi-cN=\"Archai can design your neural network with state-of-the-art neural architecture search (NAS)\">\n\t\t\t\tArchai can design your neural network with state-of-the-art neural architecture search (NAS)&nbsp;<span class=\"glyph-append glyph-append-chevron-right glyph-append-xsmall\"><\/span>\n\t\t\t<\/a>\n\t\t\t\t\t<\/li>\n\t<\/ul>\n<\/div>\n\n\n\n<p><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/archai-platform-for-neural-architecture-search\/\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>Archai<\/strong><\/a>, an open-source tool, can inform model development tradeoffs.&nbsp;In combination with a set of Neural Architecture Search (NAS) algorithms, Archai can perform a cost-aware architecture search, where \u201ccost\u201d can represent different resources of interest such as compute time or peak memory footprint. Running <em>Archai<\/em> provides the model developer with the entire spectrum of cost vs. accuracy tradeoffs, allowing them a choice of which tradeoff best meets their needs.<\/p>\n\n\n\n<p>Finally, <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/microsoft.github.io\/Accera\/\" target=\"_blank\" rel=\"noreferrer noopener\"><strong>Accera<\/strong><span class=\"sr-only\"> (opens in new tab)<\/span><\/a> is an open-source compiler that aggressively optimizes for AI workloads. The Accera compiler doesn\u2019t change or approximate a model; rather, it finds the most efficient implementation of that model. For example, matrix multiplication with a ReLU activation is commonly used in machine learning algorithms; by optimizing its implementation, developers can reduce the computational intensity of running their models.&nbsp;<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Progress in machine learning is measured in part through the constant improvement of performance metrics such as accuracy or latency. Carbon footprint metrics, while being an equally important target, have not received the same degree of attention. With contributions from our research team, Azure ML now provides transparency around machine learning resource utilization, including GPU [&hellip;]<\/p>\n","protected":false},"author":40306,"featured_media":840610,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"msr-content-parent":804847,"footnotes":""},"research-area":[],"msr-locale":[268875],"class_list":["post-847225","msr-blog-post","type-msr-blog-post","status-publish","has-post-thumbnail","hentry","msr-locale-en_us"],"msr_assoc_parent":{"id":804847,"type":"project"},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/847225"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-blog-post"}],"author":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/users\/40306"}],"version-history":[{"count":5,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/847225\/revisions"}],"predecessor-version":[{"id":870186,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-blog-post\/847225\/revisions\/870186"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/840610"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=847225"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=847225"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=847225"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}