I'm a Ph.D. student in Computer Science at UC Berkeley, where I'm part of the Berkeley Artificial Intelligence Research (BAIR) Lab and Real-Time Intelligent Secure Explainable Systems (RISE) Lab. Before moving to Berkeley, I was an undergrad at sunny UC San Diego.

Selected work

  1. Assessing generalization in deep reinforcement learning.
    Charles Packer*, Katelyn Gao*, Jernej Kos, Philipp Krähenbühl, Vladlen Koltun, Dawn Song.
    Technical Report, arXiv:1810.12282, 2019.
    summary / bibtex / code / blog

    We present a benchmark for studying generalization in deep reinforcement learning (RL). Systematic empirical evaluation shows that vanilla deep RL algorithms generalize better than specialized deep RL algorithms designed specifically for generalization. In other words, simply training on varied environments is so far the most effective strategy for generalization.

  2. Differentiable neural network architecture search.
    Richard Shin*, Charles Packer*, Dawn Song.
    ICLR Workshop, 2018.
    summary / bibtex

    The successes of deep learning in recent years has been fueled by the development of innovative new neural network architectures. We propose a method for transforming a discrete neural network architecture space into a continuous and differentiable form, which enables the use of standard gradient-based optimization techniques for this problem, and allows us to learn the architecture and the parameters simultaneously.

  3. Learning compatibility across categories for heterogeneous item recommendation.
    Ruining He, Charles Packer, Julian McAuley.
    International Conference on Data Mining (ICDM), 2016.
    summary / bibtex / data

    We propose a method for learning complex, non-metric relationships between items in a product recommendation setting. Our method, Monomer, is able to model human visual preferences by projecting image data into low-dimensional embeddings ('style' spaces).