Amy Zhang
@yayitsamyzhang
Followers
4K
Following
905
Media
3
Statuses
211
ECE prof. at UT Austin. Works in state abstractions and generalization in RL. Recruiting PhD students for Fall 26.
Joined October 2016
Check out Viraj's work on environment parallelization for online multi-task RL! ๐งต๐
Are you interested in online, multi-task RL? We built a benchmark that can run 50 Meta-World or 20 Eurekaverse tasks on a single GPU using massively parallel sim, accelerating multi-task runs to just hours and single task runs to minutes Read on for a new multi-task RL recipe๐งต
1
1
23
VERY excited for @RL_Conference next week, especially @RLBRew_RLC on Tuesday! There will be a social that evening, register here: https://t.co/tMjyxXG3OJ. Hope to see you there! @harshit_sikchi @maxbrudolph @JajooPranaya @agsidd10
luma.com
Come join us to discuss RL ideas and meet people over food and drinks (sadly not sponsored), find collaborators and friends!
1
6
39
Looking forward to #ICML2025! @ben_eysenbach and I will be giving a tutorial on RL meets GenAI tomorrow. Consider coming!
4
3
143
Unleashing the power of unordered particles: **EC-Diffuser** leverages a novel Transformer-based diffusion denoiser to learn goal-conditioned multi-object manipulation policy from pixels, achieving zero-shot generalization!๐ #ICLR2025 #diffusion #Transformers
5
14
86
๐ค Introducing RL Zero ๐ค: a new approach to transform language into behavior zero-shot for embodied agents without labeled datasets! RL Zero enables prompt-to-policy generation, and we believe this unlocks new capabilities in scaling up language-conditioned RL, providing an
3
35
214
๐ขCome to check our #NeurIPS2024 paper on how to correctly use diffusion models in RL! We rethink diffusion-based RL algorithms by using a guide-then-select paradigm to bring minimal out-of-distribution errors. Project page: https://t.co/2SkbeoI3MC
7
24
106
A new self-supervised reinforcement learning objective that learns a basis set for the optimal Q of any reward function. Check out @agsidd10โs thread below!
Do RL solutions share a common structure? We show that all solutions of Reinforcement Learning lie on a hyperplane. Our work, Proto Successor Measure, learns this abstraction of the MDP to do zero-shot RL for any reward function. (1/n)
0
2
34
Given the community response to RLC 2024, we're excited to start planning for RLC 2025. If you're interested in organizing or hosting, fill out this link:
docs.google.com
RLC (https://rl-conference.cc/) is an international conference dedicated to reinforcement learning research. This year, we have a single application for an opportunity to organize the second RLC in...
0
10
40
As part of putting together #RLC2024, we spent a bunch of time thinking about how best to improve the review process for an ML conference. Post decisions, we spent some time reflecting on what went well and what we want to change in the future: check out https://t.co/8RHmaMfKRy!
2
16
59
I did an interview on YouTube for ML New Papers. Lots of fun with @CharlesRiou007 talking about research, RL and networking in ML. Check out the video below.
#youtube #interview #reinforcementlearning #ml #machinelearning #tech I just released an interview on YouTube with ML&RL expert @yayitsamyzhang. Networking in ML, RL and its challenges, research advice... Find out below (and subscribe for more): https://t.co/7QlZZtLOUb
0
4
80
To accommodate nearby conferences and deadlines yet still maintain good reviewer matching, ๐ฐ๐ ๐๐ซ๐ ๐๐ฑ๐ญ๐๐ง๐๐ข๐ง๐ ๐ญ๐ก๐ ๐๐ฎ๐ฅ๐ฅ ๐ฉ๐๐ฉ๐๐ซ ๐๐๐๐๐ฅ๐ข๐ง๐ ๐ญ๐จ ๐๐๐ซ๐๐ก ๐๐ญ๐ก but keeping March 1st as the abstract deadline
8
17
53
A friendly reminder: two weeks until the RLC 2024 submission deadline!
4
3
15
The 3rd RLC keynote speaker is Doina Precup! She's been a major influence in so many parts of RL and we could not be more excited for her talk
0
11
94
Excited to announce the call for workshop proposals for @RL_Conference is open! https://t.co/LKIK9LCPBx Due March 8th. @JosiahHanna and I are the workshop co-chairs this year and we hope you'll consider submitting a proposal. Workshops on any RL-related topics are welcome!
0
5
21
Still using one-step dynamics model for model-based RL? Check out our work on Diffusion World Model, which is outperforming one-step models by 44% for offline D4RL tasks https://t.co/5IdwQEl9DS joint work w/ @Hanry65960814, @yayitsamyzhang @tydsh [1/3]
6
14
52
The RLC submission page is now open! Get your submissions ready: https://t.co/mQYtlnG6MY
openreview.net
Welcome to the OpenReview homepage for RLC 2024 Conference
0
19
53
Turns out offline MARL doesn't just work out of the box. Check out @paulbbarde's ๐งต๐ demonstrating why and one possible fix - which he will be presenting at @AAMASconf!
1
0
12
Check out the #NeurIPS2023 workshop on Goal Conditioned Reinforcement Learning: * Tomorrow (Friday) 900 -- 1830 CST * Great speakers: @jeffclune @r_mirsky @olexandr @ybisk @SusanMurphylab1 * Program:
0
2
17
NeurIPS RL folks, this is tonight. Come hang out with us at Howlinโ Wolf 7-10!
For folks @NeurIPSConf, we're having a social at Howling Wolf from 7 to 10 on Dec. 13th to celebrate the launch of RLC and get ideas and feedback from the community. Come by, hang out, talk about RL!
0
1
28