a large body of water with a city in the background

Advanced Machine learning, System, and Wireless Group

The goal of research on Microsoft Research Asia – Shanghai (MSRA Shanghai) is to design advanced algorithms and systems to help solve cutting-edge challenges closely related to our life. The research in this area cares about the accessing of ubiquitous signals with diverse sensing approaches, the information extraction and mining through advanced signal processing and machine learning algorithms, and technologies that builds efficient computing and sensing systems on the cloud and edge.  The algorithms and systems could help to solve the big challenges we face in healthcare, biology, and sustainability.

Research Topics

Wireless Sensing and Networking
  • Wireless sensing for next generation Human Computer Interaction (HCI)
  • Sensing and other technology for healthcare and environmental monitoring
  • 5G/6G wireless communication
Advanced Machine Learning
  • Bio-inspired neural network for general-purpose machine learning
  • Graph neural network for temporal dynamics, interpretability, and anomaly detection
  • Computer vision for vector graphics, ensemble learning, and domain adaptation/generalization
  • Sequential learning, reinforcement learning and decision making for real-world applications
  • Fundamental technologies for language generation and domain-specific NLP applications
Machine Learning for Healthcare
  • Speech recognition for patients with speech pathology
  • Baby brain seizure detection using EEG signals
  • Bio-medical knowledge graph learning for biomedicine and pharmacotherapy
  • Machine learning for transcriptomics and genomics sequences
  • Unsupervised stereotypical behavior detection for Autism
Video Streaming and Cloud Gaming Systems
  • Real-time video super resolution and frame prediction
  • Systematic optimization of video encoding, transmission, and DNN-based video enhancement
  • Server-client cooperation to mitigate bandwidth-limited and quality-unreliable network
  • Fundamental technologies of cloud gaming systems, such as job and resource scheduling
Efficient Computing for Emerging Technologies
  • AI infrastructure technologies, e.g., Kubernetes GPU schedulers and platform for deep learning workloads
  • Accelerate the training and inference of diverse models (e.g., sparse and dynamic models) on the cloud and the edge
  • Hardware efficiency (latency, energy, and carbon emission) benchmarking, prediction, and efficient model design for specific devices
  • New architecture for vector search and resource disaggregation

Want to drop by?

Unit 4301-4304 AI Tower, No.701 Yunjin Rd, Xuhui District, Shanghai, 200232, China

We’re hiring