News & features
Efficient and hardware-friendly neural architecture search with SpaceEvo
| Li Lyna Zhang, Jiahang Xu, Quanlu Zhang, Yuqing Yang, Ting Cao, and Mao Yang
A persistent challenge in deep learning is optimizing neural network models for diverse hardware configurations, balancing performance and low latency. Learn how SpaceEvo automates hardware-aware neural architecture search to fine-tune DNN models for swift execution on diverse devices.
Research Focus: Week of September 25, 2023
Chunked prefills & decode-maximal batching boost LLM inference; DragNUWA combines text, image, and trajectory for fine-grained video content control; reconstructing images from human brain signals; structural inequalities in creator-audience relationships.
Research Focus: Week of September 11, 2023
In this issue: Efficient polyglot analytics on semantic data aids query performance; generative retrieval for conversational question answering improves dialogue-based interfaces; a new tool uses ML to address capacity degradation in lithium-ion batteries.
Frontiers of multimodal learning: A responsible AI approach
New evaluation methods and a commitment to continual improvement are musts if we’re to build multimodal AI systems that advance human goals. Learn about cutting-edge research into the responsible development and use of multimodal AI at Microsoft.
Building a “heavy metal quartet” of AI compilers
A new quartet of AI compilers: Rammer, Roller, Welder, and Grinder, tackle a range of compiler optimization challenges based on the same tile abstraction, providing a comprehensive solution to connect AI models with hardware accelerators.
Research Focus: Week of August 14, 2023
In this issue: HyWay enables hybrid mingling; Auto-Tables transforms non-relational tables into standard relational forms; training dense retrievers to identify high-quality in-context examples for LLM; improving pronunciation assessment in CAPT.
Research Focus: Week of July 17, 2023
RetroRanker mitigates frequency bias in predictions of retrosynthesis models; new algorithm beats PPO on language tasks; DER dataset aids grid planning; improved PPML balances privacy & accuracy across shared data; ASL Citizen boosts sign language modeling.
Research Focus: Week of July 3, 2023
In this issue: A new study looks at engineers’ views on hybrid work; prompt engineering adds context and specificity to generative AI models; Overwatch learns code edit patterns to aid developers; a Qlib update addresses financial market challenges.
Distributional Graphormer: Toward equilibrium distribution prediction for molecular systems
| Shuxin Zheng, Chang Liu, Yu Shi, Ziheng Lu, Fusong Ju, Jianwei Zhu, Hongxia Hao, Peiran Jin, Frank Noé, Haiguang Liu, and Tie-Yan Liu
Distributional Graphormer, Microsoft’s new deep learning framework for predicting the equilibrium distribution of molecular structures, can generate realistic and diverse molecular structures with high efficiency and low cost.