Hao Li
Research Scientist
Google DeepMind

Email  /  Google Scholar  /  Github  /  Twitter

I'm a Research Scientist at Google DeepMind, working on multi-modal generative models. My research aims at developing efficient, scalable, and interpretable ML algorithms, models and systems. Preveiouly I was an Applied Scientist at Amazon AGI, where I built image/video generation and understanding models including Nova Canvas, Nova Reel, Titan Image Generator, and Custom Labels. I completed my PhD at University of Maryland, College Park, worked with Prof. Tom Goldstein and Prof. Hanan Samet on accelerating and understanding deep neural networks.

News


Selected Publications

Efficient Scaling of Diffusion Transformers for Text-to-Image Generation
Hao Li, Shamit Lal, Zhiheng Li, Yusheng Xie, Ying Wang, Yang Zou, Orchid Majumder, Aditya Golatkar, Chengwei Su, R. Manmatha, Zhuowen Tu, Stefano Ermon, Stefano Soatto, Ashwin Swaminathan
arXiv 2024 [ pdf ]
On the Scalability of Diffusion-based Text-to-Image Generation
Hao Li, Yang Zou, Ying Wang, Orchid Majumder, Yusheng Xie, R. Manmatha, Ashwin Swaminathan, Zhuowen Tu, Stefano Ermon, Stefano Soatto
CVPR 2024 [ pdf ]
Guided Recommendation for Model Fine-Tuning
Hao Li, Charless Fowlkes, Hao Yang, Onkar Dabeer, Zhuowen Tu, Stefano Soatto
CVPR 2023 [ pdf / video / poster ]
Task Adaptive Parameter Sharing for Multi-Task Learning
Matthew Wallingford, Hao Li, Alessandro Achille, Avinash Ravichandran, Charless Fowlkes, Rahul Bhotika, Stefano Soatto
CVPR 2022
Rethinking the Hyperparameters for Fine-tuning
Hao Li, Pratik Chaudhari, Hao Yang, Michael Lam, Avinash Ravichandran, Rahul Bhotika, Stefano Soatto
ICLR 2020 [video]
Visualizing the Loss Landscape of Neural Nets
Hao Li, Zheng Xu, Gavin Taylor, Christoph Studer, Tom Goldstein
NeurIPS 2018 [project / code / poster]
Training Quantized Nets: A Deeper Understanding
Hao Li*, Soham De*, Zheng Xu, Christoph Studer, Hanan Samet, Tom Goldstein
NeurIPS 2017 [poster]
ICML 2017 PADL Workshop [slides] Google Best Student Paper
Pruning Filters for Efficient ConvNets
Hao Li, Asim Kadav, Igor Durdanovic, Hanan Samet, Hans Peter Graf
ICLR 2017 [poster]

Dissertation

Towards Fast and Efficient Representation Learning
Hao Li
University of Maryland, College Park, Aug 2018 [slides]

Academic Services

  • Journal Reviewer​
    • Journal of Machine Learning Research (JMLR) (editorial board reviewer)
    • International Journal of Computer Vision (IJCV)
    • IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
    • IEEE Transactions on Neural Network and Learning Systems (TNNLS)
    • IEEE Transactions on Multimedia (TMM)
    • Neurocomputing
  • Conference Reviewer
    • NeurIPS, ICML, ICLR, ICCV, CVPR, ECCV, KDD, AAAI