Liam Li

I am a PhD student in the Machine Learning department at Carnegie Mellon University, where I work with Ameet Talwalkar. My current research focuses on automating machine learning model selection; the goal is to develop tools and algorithms that will make the practice of machine learning easier and more accessible.

Research Interests

  • automated machine learning
  • hyperparameter optimization
  • active learning
  • bandits
  • gaussian processes
  • deep learning

Publications

  • Hyperband: Bandit-Based Configuration Evaluation for Hyperparameter Optimization. L. Li, K. Jamieson, G. DeSalvo, A. Rostamizadeh, A. Talwalkar. ICLR, 2017. [PDF]
  • Hyperband: A Novel Bandit-Based Approach to Hyperparameter Optimization. L. Li, K. Jamieson, G. DeSalvo, A. Rostamizadeh, A. Talwalkar. Journal of Machine Learning Research 18 (2018) 1-52. [PDF, Blog1, Blog2]
  • Massively Parallel Hyperparameter Tuning. L. Li, K. Jamieson, A. Rostamizadeh, K. Gonina, M. Hardt, B. Recht, A. Talwalkar. In submission.
  • Pipeline-Aware Hyperparameter Optimization. L. Li, E. Sparks, R. Zhang, K. Jamieson, A. Talwalkar. In submission.

Recent Updates