Alexander C. Li

I'm a second year PhD student in the Machine Learning Department at Carnegie Mellon University. I'm advised by Deepak Pathak, and my work is supported by the National Science Foundation Graduate Research Fellowship.

Previously, I was an electrical engineering and computer science major at UC Berkeley, where I received my BS and MS. I was a researcher in the Berkeley Artificial Intelligence Research Lab and was advised by Pieter Abbeel, Lerrel Pinto, and Carlos Florensa.

Email  /  CV  /  Google Scholar  /  LinkedIn  /  Twitter

profile photo

I'm interested in self-supervised learning, generalization, optimization, and the role of data in deep learning.

Understanding Collapse in Non-Contrastive Siamese Representation Learning
Alexander C. Li, Alexei A. Efros, Deepak Pathak
ECCV 2022
pdf / code / bibtex

We empirically analyze non-contrastive self-supervised methods and find that SimSiam is extraordinarily sensitive to model size. In particular, SimSiam representations undergo partial dimensional collapse if the model is too small relative to the dataset size. We propose a metric to measure the degree of this collapse and show that it can be used to forecast the downstream task performance without any fine-tuning or labels. We further analyze architectural design choices and their effect on the downstream performance. Finally, we demonstrate that shifting to a continual learning setting acts as a regularizer and prevents collapse, and a hybrid of continual and multi-epoch training can improve linear probe accuracy by up to 18 percentage points with ResNet-18 on ImageNet.

Functional Regularization for Reinforcement Learning via Learned Fourier Features
Alexander C. Li, Deepak Pathak
NeurIPS 2021
arxiv / website / bibtex

We propose a simple architecture for deep reinforcement learning that can control how quickly the network fits different frequencies in the training data. We explain this behavior using infinite-width analysis with the Neural Tangent Kernel, and use this to prioritize learning low-frequency functions and speed up learning by reducing networks' susceptibility to noise in the optimization process, such as during Bellman updates. Experiments on standard state-based and image-based RL benchmarks show clear sample-efficiency gains, as well as increased robustness to added bootstrap noise.

Generalized Hindsight for Reinforcement Learning
Alexander C. Li, Lerrel Pinto, Pieter Abbeel
NeurIPS 2020
arxiv / project page / bibtex

We present Generalized Hindsight: an approximate inverse reinforcement learning technique for relabeling behaviors with the right tasks. Given a behavior generated under one task, Generalized Hindsight finds a different task that the behavior is better suited for. Relabeling a trajectory with this different task and training with an off-policy RL algorithm improves sample-efficiency and asymptotic performance on a suite of multi-task navigation and manipulation tasks.

Sub-policy Adaptation for Hierarchical Reinforcement Learning
Alexander C. Li*, Carlos Florensa*, Ignasi Clavera, Pieter Abbeel
International Conference on Learning Representations (ICLR), 2020
arxiv / project page / bibtex

We develop a new hierarchical RL algorithm that can efficiently adapt pre-trained skills on related tasks, and directly learn effective emergent skills by simultaneously training the entire hierarchy.

Autoregressive Models: What Are They Good For?
Murtaza Dalal*, Alexander C. Li*, Rohan Taori*
Workshop on Information Theory and Machine Learning, NeurIPS, 2019
arxiv / bibtex

We attempt to use powerful autoregressive models for image-to-image translation and outlier detection, with poor results. We analyze the failure modes and find that (a) differentiating through these models creates an ill-conditioned optimization problem and (b) their density estimates are incredibly sensitive to distributional shifts.

Sunspot Rotation and the M-Class Flare in Solar Active Region NOAA 11158
Alexander C. Li & Yang Liu
Solar Physics, 2015
paper / bibtex

We develop a novel algorithm for estimating sunspot rotation speed that is more robust to feature evolution than traditional methods based on time-slices. We use this technique to gain more accurate estimates of magnetic energy buildup in rotating sunspots.

berkeley Head TA, CS 294-158: Deep Unsupervised Learning, Spring 2020

Head Content TA, EECS 126: Stochastic Processes, Fall 2019

TA, CS 188: Artificial Intelligence, Spring 2019

TA, CS 188: Artificial Intelligence, Fall 2018

Academic Intern, CS 189: Machine Learning, Spring 2018

Reader, CS 70: Discrete Mathematics & Probability, Fall 2017
Mentor, Google Code Corps 2017
Awards and Honors

Website template from Jon Barron.