Google Scholar | Twitter

For a quick introduction to my research, see the following talk videos:

**Generative Modelling**(Slides) (Poster)**:**Implicit probabilistic models like generative adversarial nets (GANs) and variational autoencoders (VAEs) have gained popularity in recent years and have delivered impressive advances in performance. While offering substantially more modelling flexibility than prescribed probabilistic models, implicit models in general induce intractable likelihood functions and therefore cannot be trained using maximum likelihood. On the other hand, alternative training objectives have known biases; for example, GANs suffer from the well-known issues of mode collapse/dropping, vanishing gradients and training instability, which could lead to a failure to learn the underlying data distribution. We developed a method that simultaneously overcomes all three issues and show equivalence to maximizing a lower bound on log-likelihood whose tightness increases as the model capacity increases, despite not requiring the evaluation of the likelihood itself or any derived quantities. This makes it possible to train implicit probabilistic models with a likelihood-based objective, which was not previously possible using GANs or VAEs.

Related papers: Implicit Maximum Likelihood Estimation | Super-Resolution via Conditional IMLE | Diverse Image Synthesis from Semantic Layouts via Conditional IMLE | On the Implicit Assumptions of GANs

**Learning to Optimize**(Slides) (Poster)**:**While machine learning has been applied to a wide range of domains, one domain that has conspicuously been left untouched is the design of tools that power machine learning itself. In this line of work, we ask the following question: is it possible to automate the design of algorithms used in machine learning? We introduced the first framework for learning a general-purpose iterative optimization algorithm automatically. The key idea is to treat the design of an optimization algorithm as a reinforcement learning/optimal control problem and view a particular update formula (and therefore a particular optimization algorithm) as a particular policy. Finding the optimal policy then corresponds to finding the best optimization algorithm. We parameterize the update formula using a neural net and train it using reinforcement learning to avoid the problem of compounding errors. This has inspired various subsequent work on meta-learning.

Related papers: Learning to Optimize | Learning to Optimize Neural Nets

**Fast Nearest Neighbour Search**(Slides) (Poster)**:**The method of*k*-nearest neighbours is widely used in machine learning, statistics, bioinformatics and database systems. Attempts at devising fast algorithms, however, have come up against a recurring obstacle: the curse of dimensionality. Almost all exact algorithms developed over the past 40 years exhibited a time complexity that is exponential in ambient or intrinsic dimensionality, and such persistent failure in overcoming the curse of dimensionality led to conjectures that doing so is impossible. We showed that, surprisingly, this is in fact possible — we developed an exact randomized algorithm whose query time complexity is linear in ambient dimensionality and sublinear in intrinsic dimensionality. The key insight is to avoid the popular strategy of space partitioning, which we argue gives rise to the curse of dimensionality. We demonstrated a speedup of 1-2 orders of magnitude over locality-sensitive hashing (LSH).

Related papers: Fast*k*-Nearest Neighbour Search via Dynamic Continuous Indexing | Fast*k*-Nearest Neighbour Search via Prioritized DCI

- Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Ning Yu,**Ke Li**, Peng Zhou, Jitendra Malik, Larry Davis, Mario Fritz

*European Conference on Computer Vision (ECCV)*, 2020 - Multimodal Image Synthesis with Conditional Implicit Maximum Likelihood Estimation

**Ke Li***, Shichong Peng*, Tianhao Zhang*, Jitendra Malik

*International Journal of Computer Vision (IJCV)*, 2019 - Diverse Image Synthesis from Semantic Layouts via Conditional IMLE (Project Page) (Code) (Talk)

**Ke Li***, Tianhao Zhang*, Jitendra Malik

*IEEE International Conference on Computer Vision (ICCV)*, 2019 - Non-Adversarial Image Synthesis with Generative Latent Nearest Neighbors (Code) (Talk)

Yedid Hoshen,**Ke Li**, Jitendra Malik

*IEEE Conference on Computer Vision and Pattern Recognition (CVPR)*, 2019 - On the Implicit Assumptions of GANs (Poster)

**Ke Li**, Jitendra Malik

*NeurIPS Workshop on Critiquing and Correcting Trends in Machine Learning*, 2018 - Super-Resolution via Conditional Implicit Maximum Likelihood Estimation (Project Page) (Talk)

**Ke Li***, Shichong Peng*, Jitendra Malik

*arXiv:1810.01406*, 2018 - Implicit Maximum Likelihood Estimation (Project Page) (Reviews) (Slides) (Poster) (Code) (Talk)

**Ke Li**, Jitendra Malik

*arXiv:1809.09087*, 2018

- Learning to Optimize Neural Nets (Slides) (Blog Post) (Talk)

**Ke Li**, Jitendra Malik

*arXiv:1703.00441*, 2017 - Learning to Optimize (ICLR Version) (Slides) (Poster) (Code) (Blog Post) (Talk)

**Ke Li**, Jitendra Malik

*arXiv:1606.01885*, 2016 and*International Conference on Learning Representations (ICLR)*, 2017

- Fast
*k*-Nearest Neighbour Search via Prioritized DCI (Talk) (Slides) (Project Page) (Code) (Poster)

**Ke Li**, Jitendra Malik

*International Conference on Machine Learning (ICML)*, 2017 - Fast
*k*-Nearest Neighbour Search via Dynamic Continuous Indexing (Slides) (Project Page) (Code)

**Ke Li**, Jitendra Malik

*International Conference on Machine Learning (ICML)*, 2016

- Amodal Instance Segmentation

**Ke Li**, Jitendra Malik

*European Conference on Computer Vision (ECCV)*, 2016 - Iterative Instance Segmentation

**Ke Li**, Bharath Hariharan, Jitendra Malik

*IEEE Conference on Computer Vision and Pattern Recognition (CVPR)*, 2016

- Approximate Feature Collisions in Neural Nets

**Ke Li***, Tianhao Zhang*, Jitendra Malik

*Advances in Neural Information Processing Systems (NeurIPS)*, 2019 - Trajectory Normalized Gradients for Distributed Optimization

Jianqiao Wangni,**Ke Li**, Jianbo Shi, Jitendra Malik

*arXiv:1901.08227*, 2019 - Are All Training Examples Created Equal? An Empirical Study

Kailas Vodrahalli,**Ke Li**, Jitendra Malik

*arXiv:1811.12569*, 2018 - Efficient Feature Learning using Perturb-and-MAP

**Ke Li**, Kevin Swersky, Richard Zemel

*NIPS Workshop on Perturbations, Optimization and Statistics*, 2013

- University of Illinois at Urbana-Champaign — Feb 2020
- University of Washington — Jan 2020
- University of Texas at Austin — Jan 2020
- Vector Institute for Artificial Intelligence — Dec 2019
- Stanford University — Dec 2019
- Institute for Advanced Study (IAS) — Oct 2019
- Google NYC — Oct 2019
- Massachusetts Institute of Technology — Oct 2019
- Cornell Tech — Oct 2019
- Carnegie Mellon University — Oct 2019
- Simons Institute for the Theory of Computing — Jun 2019
- Google Seattle — Jun 2019
- DeepMind — Jun 2019
- University of California, Berkeley — May 2019

- Nvidia — Dec 2019
- Google Mountain View — Dec 2019
- BAIR/BDD Computer Vision Workshop — Sep 2019
- Adobe — Aug 2019
- Nielsen — Jul 2019

- BAIR/FAIR Workshop — Aug 2019
- University of California, Berkeley — Aug 2018

- BAIR Seminar — Aug 2019
- CIFAR Deep Learning and Reinforcement Learning Summer School (DLRLSS) — Jul 2019

- Google NYC — Jan 2020
- Simons Institute for the Theory of Computing — Nov 2018
- NIPS 2017 Workshop on Nearest Neighbours for Modern Applications with Massive Data — Dec 2017

- Institute for Advanced Study (IAS) — Apr 2020

- BAIR Fall Workshop — Oct 2017
- University of Toronto — Jun 2017
- University of California, Berkeley — Feb 2017

- Intuition Machines Seminar — Apr 2017

- IEEE Transactions on Information Theory
- IEEE Transactions on Signal Processing
- IEEE Transactions on Neural Networks and Learning Systems
- Information Sciences

- NeurIPS, ICML, ICLR, CVPR, ICCV, ECCV, AAAI