{"id":881388,"date":"2022-09-27T19:41:19","date_gmt":"2022-09-28T02:41:19","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/"},"modified":"2024-07-02T02:00:36","modified_gmt":"2024-07-02T09:00:36","slug":"sh-sys-eng-group","status":"publish","type":"msr-group","link":"https:\/\/www.microsoft.com\/en-us\/research\/group\/sh-sys-eng-group\/","title":{"rendered":"MSRA (Shanghai) System and Engineering Group"},"content":{"rendered":"
\n\t
\n\t\t
\n\t\t\t\"View\t\t<\/div>\n\t\t\n\t\t
\n\t\t\t\n\t\t\t
\n\t\t\t\t\n\t\t\t\t
\n\t\t\t\t\t\n\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\n\t\t\t\t\t\t\t\n\n

System and Engineering Group<\/h1>\n\n\n\n

MSR Asia-Shanghai<\/p>\n\n\t\t\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\t<\/div>\n<\/section>\n\n\n\n\n\n

The System and Engineering group<\/strong> of Microsoft Research Asia (Shanghai) is a pioneering force in the realm of deep learning systems and their integration into the expansive landscape of large language models (LLMs) and artificial intelligence (AI) ecosystems. Our research spans multiple dimensions, from developing efficient inference engines that harness the power of sparsity and dynamism to advancing AI infrastructure technologies and exploring emerging applications for LLMs. We are committed to optimizing computing for emerging technologies, reducing hardware inefficiencies, and designing new architectures. In addition, our expertise extends to real-time video enhancement and cloud gaming systems, ensuring superior quality and reliability in multimedia experiences. As we navigate this ever-evolving field, our group remains at the forefront, shaping the future of AI systems and infrastructure.<\/p>\n\n\n\n

\u5fae\u8f6f\u4e9a\u6d32\u7814\u7a76\u9662\uff08\u4e0a\u6d77\uff09\u7cfb\u7edf\u4e0e\u5de5\u7a0b\u7ec4<\/strong>\u6df1\u8015\u5728\u6df1\u5ea6\u5b66\u4e60\u7cfb\u7edf\u9886\u57df\uff0c\u4e13\u6ce8\u4e8e\u5c06\u5176\u878d\u5165\u5927\u578b\u8bed\u8a00\u6a21\u578b\uff08LLMs\uff09\u548c\u4eba\u5de5\u667a\u80fd\uff08AI\uff09\u751f\u6001\u7cfb\u7edf\u7684\u5e7f\u6cdb\u9886\u57df\u3002\u6211\u4eec\u7684\u7814\u7a76\u6db5\u76d6\u591a\u4e2a\u65b9\u9762\uff0c\u4ece\u5f00\u53d1\u9ad8\u6548\u7684\u63a8\u7406\u5f15\u64ce\uff0c\u5145\u5206\u53d1\u6325\u7a00\u758f\u6027\u548c\u52a8\u6001\u6027\u7684\u6f5c\u529b\uff0c\u5230\u63a8\u52a8AI\u57fa\u7840\u8bbe\u65bd\u6280\u672f\u7684\u8fdb\u6b65\uff0c\u4ee5\u53ca\u63a2\u7d22LLMs\u7684\u65b0\u5174\u5e94\u7528\u3002\u6211\u4eec\u81f4\u529b\u4e8e\u4f18\u5316\u65b0\u5174\u6280\u672f\u7684\u8ba1\u7b97\uff0c\u51cf\u5c11\u786c\u4ef6\u6548\u7387\u95ee\u9898\uff0c\u8bbe\u8ba1\u5168\u65b0\u7684\u67b6\u6784\u3002\u6b64\u5916\uff0c\u6211\u4eec\u7684\u4e13\u4e1a\u9886\u57df\u8fd8\u6db5\u76d6\u4e86\u5b9e\u65f6\u89c6\u9891\u589e\u5f3a\u548c\u4e91\u6e38\u620f\u7cfb\u7edf\uff0c\u4ee5\u786e\u4fdd\u591a\u5a92\u4f53\u4f53\u9a8c\u7684\u5353\u8d8a\u54c1\u8d28\u548c\u53ef\u9760\u6027\u3002\u5728\u8fd9\u4e2a\u4e0d\u65ad\u6f14\u5316\u7684\u9886\u57df\u4e2d\uff0c\u6211\u4eec\u7684\u56e2\u961f\u4e00\u76f4\u7ad9\u5728\u524d\u6cbf\uff0c\u5851\u9020\u7740AI\u7cfb\u7edf\u548c\u57fa\u7840\u8bbe\u65bd\u7684\u672a\u6765\u3002<\/p>\n\n\n\n

Research topics<\/h2>\n\n\n\n
System for Large Language Models and Ecosystem<\/h5>\n\n\n\n
    \n
  • Efficient inference engine by leveraging sparsity and dynamism from model architecture, values, and inputs<\/li>\n\n\n\n
  • AI infrastructure technologies, e.g., Kubernetes GPU schedulers and platform for deep learning workloads<\/li>\n\n\n\n
  • Emerging technologies for LLM applications including Copilots and Autonomous Agents, such as prompt compression, and lifelong learning from historical records<\/li>\n\n\n\n
  • Key components connected with LLM, such as data service and vector search<\/li>\n<\/ul>\n\n\n\n
    Efficient Computing for Emerging Technologies<\/h5>\n\n\n\n
      \n
    • Accelerate the training and inference of diverse models on the cloud and the edge<\/li>\n\n\n\n
    • Hardware efficiency (latency, energy, and carbon emission) benchmarking, prediction, and efficient model design for specific devices<\/li>\n\n\n\n
    • New architecture for vector search and resource disaggregation<\/li>\n<\/ul>\n\n\n\n
      Video Streaming and Cloud Gaming Systems<\/h5>\n\n\n\n
        \n
      • Real-time video super resolution and frame prediction<\/li>\n\n\n\n
      • Systematic optimization of video encoding, transmission, and DNN-based video enhancement<\/li>\n\n\n\n
      • Server-client cooperation to mitigate bandwidth-limited and quality-unreliable network<\/li>\n\n\n\n
      • Fundamental technologies of cloud gaming systems, such as job and resource scheduling<\/li>\n<\/ul>\n\n\n","protected":false},"excerpt":{"rendered":"

        MSR Asia-Shanghai The System and Engineering group of Microsoft Research Asia (Shanghai) is a pioneering force in the realm of deep learning systems and their integration into the expansive landscape of large language models (LLMs) and artificial intelligence (AI) ecosystems. Our research spans multiple dimensions, from developing efficient inference engines that harness the power of […]<\/p>\n","protected":false},"featured_media":830866,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"msr_group_start":"","footnotes":""},"research-area":[13556,13547],"msr-group-type":[243694],"msr-locale":[268875],"msr-impact-theme":[],"class_list":["post-881388","msr-group","type-msr-group","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-research-area-systems-and-networking","msr-group-type-group","msr-locale-en_us"],"msr_group_start":"","msr_detailed_description":"","msr_further_details":"","msr_hero_images":[],"msr_research_lab":[],"related-researchers":[{"type":"user_nicename","display_name":"Yuqing Yang","user_id":40654,"people_section":"Section name 0","alias":"yuqyang"},{"type":"user_nicename","display_name":"Nan Chen","user_id":42033,"people_section":"Section name 0","alias":"nanchen"},{"type":"user_nicename","display_name":"Zhiyuan He","user_id":41743,"people_section":"Section name 0","alias":"zhiyuhe"},{"type":"user_nicename","display_name":"Huiqiang Jiang","user_id":40807,"people_section":"Section name 0","alias":"hjiang"},{"type":"user_nicename","display_name":"Luna K. Qiu","user_id":41353,"people_section":"Section name 0","alias":"lunaqiu"},{"type":"user_nicename","display_name":"Guoxin Sui","user_id":41566,"people_section":"Section name 0","alias":"gusui"},{"type":"user_nicename","display_name":"Jiahang Xu","user_id":41569,"people_section":"Section name 0","alias":"jiahangxu"},{"type":"user_nicename","display_name":"Yifan Yang","user_id":41539,"people_section":"Section name 0","alias":"yifanyang"},{"type":"user_nicename","display_name":"Yuge Zhang","user_id":41659,"people_section":"Section name 0","alias":"yugzhan"},{"type":"user_nicename","display_name":"Chengruidong Zhang","user_id":42018,"people_section":"Section name 0","alias":"chengzhang"},{"type":"user_nicename","display_name":"Siyun Zhao","user_id":42009,"people_section":"Section name 0","alias":"siyunzhao"}],"related-publications":[858315,964458,1053816,885912,964464,1053831,887244,964470,910137,974562,666816,941055,975066,700303,945174,978312,700309,946797,981918,748264,954759,990306,847507,954765,999000,854772,955587,999015,858267,962919,1016619],"related-downloads":[1057272],"related-videos":[],"related-projects":[],"related-events":[],"related-opportunities":[],"related-posts":[],"tab-content":[],"msr_impact_theme":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/881388"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-group"}],"version-history":[{"count":13,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/881388\/revisions"}],"predecessor-version":[{"id":1076349,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group\/881388\/revisions\/1076349"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/830866"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=881388"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=881388"},{"taxonomy":"msr-group-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-group-type?post=881388"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=881388"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=881388"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}