PhD student @rlai_lab. Prev: @GoogleDeepMind, @OcadoTechnology, @kindredai, @UofTCompSci
Edmonton
Joined October 2020
- Tweets52
- Following546
- Followers196
- Likes4.2K
Bryan Chan retweeted
Meet the recipients of the 2024 ACM A.M. Turing Award, Andrew G. Barto and Richard S. Sutton! They are recognized for developing the conceptual and algorithmic foundations of reinforcement learning. Please join us in congratulating the two recipients! bit.ly/4hpdsbD
Bryan Chan retweeted
Would you believe that deep RL can work without replay buffers, target networks, or batch updates? Our recent work gets deep RL agents to learn from a continuous stream of data one sample at a time without storing any sample. Joint work with @Gautham529 and @rupammahmood.
Bryan Chan retweeted
Our NeurIPS paper is now on arXiv:
We introduce Action Value Gradient (AVG), a novel incremental deep RL method that learns in real-time, one sample at a time — no batch updates, target networks or a replay buffer!
Co-authors @mhmd_elsaye @bellingerc @white_martha @rupammahmood
Bryan Chan retweeted
Check out our (myself and @chanpyb) recently accepted @ieeeras RA-L version of Learning from Guided Play!
We show that simple auxiliary tasks can significantly improve exploration, and resolve deceptive rewards found with Adversarial Imitation Learning.
papers.starslab.ca/lfgp
Bryan Chan retweeted
Inverse RL is susceptible to learning deceptive rewards and failing at demonstrated tasks, especially when applied to manipulators. Our latest work on Learning from Guided Play, led by @trevorablett, has been accepted to @ieeeras RA-L!
Preprint:
arxiv.org/abs/2301.00051
Bryan Chan retweeted
Enough games. The RL field needs to mature.
New blog post with @shiemannor
avivtamar.substack.com/p/dep…
Bryan Chan retweeted
Allow your experts to play...where have I seen this before??
papers.starslab.ca/lfgp/ @utiasSTARS @chanpyb
Bryan Chan retweeted
Robotic Imitation Learning fans! Have you found that Adversarial IL performs poorly on manipulation tasks? Us too!
Learning from Guided Play (LfGP), accepted at the 2021 Neurips Deep RL workshop, addresses this with simple auxiliary tasks.
Links in 🧵👇, work with @chanpyb
Bryan Chan retweeted
We are excited to launch a pilot Graduate Application Assistance Program @UofTCompSci!
Current graduate students help review pre-submission application materials with the focus on guiding underrepresented applicants.
Details: sites.google.com/view/toront…
Bryan Chan retweeted
PhD candidate Filip Marić is an invited speaker at the @RoboticsSciSys Workshop on Geometry and Topology in Robotics! He will discuss how advances in computational geometry open up new perspectives for reasoning about robot kinematics.
Check out his talk:
youtu.be/a2pG11CV_ug