ICML 2020 Oral Talk Collection

I found a great collection of ICML 2020 paper presentation videos shared by the authors. You can access it here without the conference registration.

Here are some talks I found interesting:

• Generative Pretraining from Pixels Review (Outstanding Paper HM, by OpenAI)

• Estimating Q (s, s') with Deep Deterministic Dynamics Gradients (by Uber AI & Georgia Tech)

• Estimation of Bounds on Potential Outcomes For Decision Making (by MIT)

• Understanding and Mitigating the Tradeoff between Robustness and Accuracy (by Stanford & ETH Zurich)

• Planning to explore via self-supervised world models (by UPenn, Berkeley, Google, FB, CMU, etc)

Hope to see more talks coming out soon...