René's URL Explorer Experiment


Title: psc's website

Description: Homepage of Pablo Samuel Castro

Generator: Hugo 0.163.3

direct link

Domain: psc-g.github.io

Noneie=edge

Links:

psc's websitehttps://psc-g.github.io/
Homehttps://psc-g.github.io/#home
Abouthttps://psc-g.github.io/#about
Recent Postshttps://psc-g.github.io/#recent-posts
Selected Publicationshttps://psc-g.github.io/#publications
Studentshttps://psc-g.github.io/#students
Postshttps://psc-g.github.io/posts
https://psc-g.github.io/#about
Googlehttps://ai.google/
public office hourshttps://calendar.app.google/SRAa924Z2ac2D31x9
https://twitter.com/pcastr
https://github.com/psc-g
In Defense of Atari - the ALE is not 'solved'!This post is based on a talk I gave at the AutoRL workshop in ICML 2024, which unfortunately was not recorded.IntroductionReinforcement Learning (RL) has been used successfully in a number of challenging tasks, such as beating world champions at Go, controlling tokamak plasmas for nuclear fusion, optimized chip placement, and controlling stratospheric balloons. All these successes have leveraged years of research and expertise and, importantly, rely on the combination of RL algorithms with deep neural networks (as proposed in the seminal DQN paper).December 2, 2024 Readhttps://psc-g.github.io/posts/research/rl/atari_defense/
AutoRL workshop in ICML 2024https://autorlworkshop.github.io/
beating world champions at Gohttps://deepmind.google/research/breakthroughs/alphago/
controlling tokamak plasmas for nuclear fusionhttps://www.nature.com/articles/s41586-021-04301-9
optimized chip placementhttps://www.nature.com/articles/s41586-021-03544-w
controlling stratospheric balloonshttps://www.nature.com/articles/s41586-020-2939-8
DQN paperhttps://www.nature.com/articles/nature14236
Readhttps://psc-g.github.io/posts/research/rl/atari_defense/
From "Bigger, Better, Faster" to "Smaller, Sparser, Stranger"This is a post based on a talk I gave a few times in 2023. I had been meaning to put it in blog post form for over a year but kept putting it off… I guess better late than never. I think some of the ideas still hold, so hope some of you find it useful!Bigger, better, fasterIn the seminal DQN paper, Mnih et al. demonstrated that reinforcement learning, when combined with neural networks as function approximators, could learn to play Atari 2600 games at superhuman levels. The DQN agent learned to do this over 200 million environment frames, which is roughly equivalent to 1000 hours of human gameplay…November 27, 2024 Readhttps://psc-g.github.io/posts/research/rl/from_bbf_to_sss/
seminal DQN paperhttps://www.nature.com/articles/nature14236
Readhttps://psc-g.github.io/posts/research/rl/from_bbf_to_sss/
The Dormant Neuron Phenomenon in Deep Reinforcement LearningWe identify the dormant neuron phenomenon in deep reinforcement learning, where an agent’s network suffers from an increasing number of inactive neurons, thereby affecting network expressivity.Ghada Sokar, Rishabh Agarwal, Pablo Samuel Castro*, Utku Evci*This blogpost is a summary of our ICML 2023 paper. The code is available here. Many more results and analyses are available in the paper, so I encouraged you to check it out if interested!The following figure gives a nice summary of the overall findings of our work (we are reporting the Interquantile Mean (IQM) as introduced in our Statistical Precipice NeurIPS'21 paper):June 19, 2023 Readhttps://psc-g.github.io/posts/research/rl/redo/
ICML 2023 paperhttps://arxiv.org/abs/2302.12902
herehttps://github.com/google/dopamine/tree/master/dopamine/labs/redo
our Statistical Precipice NeurIPS'21 paperhttps://arxiv.org/abs/2108.13264
Readhttps://psc-g.github.io/posts/research/rl/redo/
Simplicial Embeddings Improve Sample Efficiency in Actor-Critic AgentsJohan Obando-Ceron, Walter Mayor, Samuel Lavoie, Scott Fujimoto, Aaron Courville, Pablo Samuel Castro ICLR, 2026https://openreview.net/forum?id=mCpq1GCKxA
Paperhttps://openreview.net/forum?id=mCpq1GCKxA
ARM-FM - Automated Reward Machines via Foundation Models for Compositional Reinforcement LearningRoger Creus Castanyer, Faisal Mohamed, Pablo Samuel Castro, Cyrus Neary, Glen Berseth ICLR, 2026https://openreview.net/forum?id=OBpQdCWLfd
Paperhttps://openreview.net/forum?id=OBpQdCWLfd
Discovering Differences in Strategic Behavior between Humans and LLMsCaroline Wang, Daniel Kasenberg, Kim Stachenfeld, Pablo Samuel Castro ICML, 2026https://openreview.net/forum?id=Y9RqhWmcyK
Paperhttps://openreview.net/forum?id=Y9RqhWmcyK
Stable Deep Reinforcement Learning via Isotropic Gaussian RepresentationsAli Saheb Pasand, Johan Obando-Ceron, Aaron Courville, Pouya Bashivan, Pablo Samuel Castro ICML, 2026https://openreview.net/forum?id=gc7Gg18ejz
Paperhttps://openreview.net/forum?id=gc7Gg18ejz
Stable Gradients for Stable Learning at Scale in Deep Reinforcement LearningRoger Creus Castanyer, Johan Obando-Ceron, Lu Li, Pierre-Luc Bacon, Glen Berseth, Aaron Courville, Pablo Samuel Castro NeurIPS, 2025https://openreview.net/forum?id=Vqj65VeDOu
Paperhttps://openreview.net/forum?id=Vqj65VeDOu
Don't flatten, tokenize! Unlocking the key to SoftMoE's efficacy in deep RLGhada Sokar, Johan Obando-Ceron, Aaron Courville, Hugo Larochelle, Pablo Samuel Castro ICLR, 2025https://proceedings.iclr.cc/paper_files/paper/2025/hash/74d056577c164d45f713ce297a4872df-Abstract-Conference.html
Paperhttps://proceedings.iclr.cc/paper_files/paper/2025/hash/74d056577c164d45f713ce297a4872df-Abstract-Conference.html
Mind the GAP! The Challenges of Scale in Pixel-based Deep Reinforcement LearningGhada Sokar, Pablo Samuel Castro NeurIPS, 2025https://openreview.net/forum?id=LrBWGwVfCA
Paperhttps://openreview.net/forum?id=LrBWGwVfCA
Meta-World+ - An Improved, Standardized, RL BenchmarkReginald McLean, Evangelos Chatzaroulas, Luc McCutcheon, Frank Röder, Tianhe Yu, Zhanpeng He, K.R. Zentner, Ryan Julian, J K Terry, Isaac Woungang, Nariman Farsad, Pablo Samuel Castro NeurIPS D&B, 2025https://openreview.net/forum?id=1de3azE606
Paperhttps://openreview.net/forum?id=1de3azE606
Optimistic critics can empower small actorsOlya Mastikhina, Dhruv Sreenivas, Pablo Samuel Castro RLC, 2025https://openreview.net/forum?id=yx68Ns2e5E
Paperhttps://openreview.net/forum?id=yx68Ns2e5E
Multi-Task Reinforcement Learning Enables Parameter ScalingReginald McLean, Evangelos Chatzaroulas, Jordan Terry, Isaac Woungang, Nariman Farsad, Pablo Samuel Castro RLC, 2025https://openreview.net/forum?id=eBWwBIFV7T#discussion
Paperhttps://openreview.net/forum?id=eBWwBIFV7T#discussion
Discovering Symbolic Cognitive Models from Human and Animal BehaviorPablo Samuel Castro, Nenad Tomasev, Ankit Anand, Navodita Sharma, Rishika Mohanta, Aparna Dev, Kuba Perlin, Siddhant Jain, Kyle Levin, Noemi Elteto, Will Dabney, Alexander Novikov, Glenn C Turner, Maria K Eckstein, Nathaniel D. Daw, Kevin J Miller, Kim Stachenfeld ICML, 2025https://openreview.net/forum?id=dhRXGWJ027
Paperhttps://openreview.net/forum?id=dhRXGWJ027
The Impact of On-Policy Parallelized Data Collection on Deep Reinforcement Learning NetworksWalter Mayor, Johan Obando-Ceron, Aaron Courville, Pablo Samuel Castro ICML, 2025https://openreview.net/forum?id=cnqyzuZhSo
Paperhttps://openreview.net/forum?id=cnqyzuZhSo
A Survey of State Representation Learning for Deep Reinforcement LearningAyoub Echchahed, Pablo Samuel Castro TMLR, 2025https://openreview.net/forum?id=gOk34vUHtz
Paperhttps://openreview.net/forum?id=gOk34vUHtz
CALE - Continuous Arcade Learning EnvironmentJesse Farebrother, Pablo Samuel Castro NeurIPS, 2024https://psc-g.github.io/posts/research/rl/atari_defense
GitHubhttps://github.com/Farama-Foundation/Arcade-Learning-Environment
Paperhttps://arxiv.org/abs/2410.23810
Blog posthttps://psc-g.github.io/posts/research/rl/atari_defense
On the consistency of hyper-parameter selection in value-based deep reinforcement learningJohan Obando-Ceron, João G.M. Araújo, Aaron Courville, Pablo Samuel Castro RLC, 2024https://psc-g.github.io/posts/research/rl/atari_defense
GitHubhttps://consistent-hparams.streamlit.app/
Paperhttps://arxiv.org/abs/2406.17523
Blog posthttps://psc-g.github.io/posts/research/rl/atari_defense
Mixture of Experts in a Mixture of RL settingsTimon Willi, Johan Obando-Ceron, Jakob Foerster, Karolina Dziugaite, Pablo Samuel Castro RLC, 2024https://arxiv.org/abs/2406.18420
Paperhttps://arxiv.org/abs/2406.18420
In value-based deep reinforcement learning, a pruned network is a good networkJohan Obando-Ceron, Aaron Courville, Pablo Samuel Castro ICML, 2024https://arxiv.org/abs/2402.12479
GitHubhttps://github.com/google/dopamine/
Paperhttps://arxiv.org/abs/2402.12479
Mixtures of experts unlock parameter scaling for deep RLJohan Obando-Ceron, Ghada Sokar, Timon Willi, Clare Lyle, Jesse Farebrother, Jakob Foerster, Gintare Karolina Dziugaite, Doina Precup, Pablo Samuel Castro ICML, 2024https://arxiv.org/abs/2402.08609
GitHubhttps://github.com/google/dopamine/tree/master/dopamine/labs/moes
Paperhttps://arxiv.org/abs/2402.08609
Stop Regressing - Training Value Functions via Classification for Scalable Deep RLJesse Farebrother, Jordi Orbay, Quan Vuong, Adrien Ali Taïga, Yevgen Chebotar, Ted Xiao, Alex Irpan, Sergey Levine, Pablo Samuel Castro, Aleksandra Faust, Aviral Kumar, Rishabh Agarwal ICML, 2024https://arxiv.org/abs/2403.03950
Paperhttps://arxiv.org/abs/2403.03950
Small batch deep reinforcement learningJohan Obando Ceron, Marc Bellemare, Pablo Samuel Castro NeurIPS, 2023https://proceedings.neurips.cc/paper_files/paper/2023/hash/528388f1ad3a481249a97cbb698d2fe6-Abstract-Conference.html
GitHubhttps://github.com/google/dopamine
Paperhttps://proceedings.neurips.cc/paper_files/paper/2023/hash/528388f1ad3a481249a97cbb698d2fe6-Abstract-Conference.html
Minigrid & miniworld - Modular & customizable reinforcement learning environments for goal-oriented tasksMaxime Chevalier-Boisvert, Bolun Dai, Mark Towers, Rodrigo Perez-Vicente, Lucas Willems, Salem Lahlou, Suman Pal, Pablo Samuel Castro, Jordan Terry NeurIPS, 2023https://proceedings.neurips.cc/paper_files/paper/2023/hash/e8916198466e8ef218a2185a491b49fa-Abstract-Datasets_and_Benchmarks.html
GitHubhttps://minigrid.farama.org/
Paperhttps://proceedings.neurips.cc/paper_files/paper/2023/hash/e8916198466e8ef218a2185a491b49fa-Abstract-Datasets_and_Benchmarks.html
A Kernel Perspective on Behavioural Metrics for Markov Decision ProcessesPablo Samuel Castro, Tyler Kastner, Prakash Panangaden, Mark Rowland TMLR, 2023https://openreview.net/forum?id=nHfPXl1ly7
GitHubhttps://github.com/google-research/google-research/tree/master/ksme
Paperhttps://openreview.net/forum?id=nHfPXl1ly7
Bigger, Better, Faster, Human-level Atari with human-level efficiencyMax Schwarzer, Johan Obando-Ceron, Aaron Courville, Marc Bellemare, Rishabh Agarwal, Pablo Samuel Castro ICML, 2023https://psc-g.github.io/posts/research/rl/from_bbf_to_sss
GitHubhttps://github.com/google-research/google-research/tree/master/bigger_better_faster
Paperhttps://arxiv.org/abs/2305.19452
Blog posthttps://psc-g.github.io/posts/research/rl/from_bbf_to_sss
The dormant neuron phenomenon in deep reinforcement learningGhada Sokar, Rishabh Agarwal, Pablo Samuel Castro, Utku Evci ICML, 2023https://psc-g.github.io/posts/research/rl/redo
GitHubhttps://github.com/google/dopamine/tree/master/dopamine/labs/redo
Paperhttps://arxiv.org/abs/2302.12902
Blog posthttps://psc-g.github.io/posts/research/rl/redo
Proto-Value Networks, Scaling Representation Learning with Auxiliary TasksJesse Farebrother, Joshua Greaves, Rishabh Agarwal, Charline Le Lan, Ross Goroshin, Pablo Samuel Castro, Marc G. Bellemare ICLR, 2023https://openreview.net/forum?id=oGDKSt9JrZi
GitHubhttps://github.com/google-research/google-research/tree/master/pvn
Paperhttps://openreview.net/forum?id=oGDKSt9JrZi
Reincarnating Reinforcement Learning, Reusing Prior Computation to Accelerate ProgressRishabh Agarwal, Max Schwarzer, Pablo Samuel Castro, Aaron Courville, Marc G. Bellemare NeurIPS, 2022https://arxiv.org/abs/2206.01626
GitHubhttps://github.com/google-research/reincarnating_rl
Paperhttps://arxiv.org/abs/2206.01626
The State of Sparse Training in Deep Reinforcement LearningLaura Graesser*, Utku Evci*, Erich Elsen, and Pablo Samuel Castro ICML, 2022https://psc-g.github.io/posts/research/rl/sparse_rl
GitHubhttps://github.com/google-research/rigl/tree/master/rigl/rl
Paperhttps://arxiv.org/abs/2206.10369
Blog posthttps://psc-g.github.io/posts/research/rl/sparse_rl
A general class of surrogate functions for stable and efficient reinforcement learningSharan Vaswani, Olivier Bachem, Simone Totaro, Robert Mueller, Shivam Garg, Matthieu Geist, Marlos C. Machado, Pablo Samuel Castro, and Nicolas Le Roux AISTATS, 2022https://arxiv.org/abs/2108.05828
GitHubhttps://github.com/svmgrg/fma-pg
Paperhttps://arxiv.org/abs/2108.05828
MICo, Learning improved representations via sampling-based state similarity for Markov decision processesPablo Samuel Castro, Tyler Kastner, Prakash Panangaden, and Mark Rowland NeurIPS, 2021https://psc-g.github.io/posts/research/rl/mico
GitHubhttps://github.com/google-research/google-research/tree/master/mico
Paperhttps://arxiv.org/abs/2106.08229
Blog posthttps://psc-g.github.io/posts/research/rl/mico
The Difficulty of Passive Learning in Deep Reinforcement LearningGeorg Ostrovski, Pablo Samuel Castro, Will Dabney NeurIPS, 2021https://psc-g.github.io/posts/research/rl/tandem
GitHubhttps://github.com/deepmind/deepmind-research/tree/master/tandem_dqn
Paperhttps://arxiv.org/abs/2110.14020
Blog posthttps://psc-g.github.io/posts/research/rl/tandem
Deep Reinforcement Learning at the Edge of the Statistical PrecipiceRishabh Agarwal, Max Schwarzer, Pablo Samuel Castro, Aaron Courville, and Marc G. Bellemare NeurIPS, 2021https://psc-g.github.io/posts/research/rl/precipice
GitHubhttps://github.com/google-research/rliable
Paperhttps://arxiv.org/abs/2108.13264
Blog posthttps://psc-g.github.io/posts/research/rl/precipice
Losses, Dissonances, and DistortionsPablo Samuel Castro Machine Learning for Creativity and Design workshop at NeurIPS, 2021https://psc-g.github.io/posts/musicode/ldd/
GitHubhttps://github.com/psc-g/musicode/tree/main/ldd
Paperhttps://arxiv.org/abs/2111.05128
Blog posthttps://psc-g.github.io/posts/musicode/ldd/
Revisiting Rainbow, Promoting more insightful and inclusive deep reinforcement learning researchJohan S. Obando-Ceron and Pablo Samuel Castro ICML, 2021https://psc-g.github.io/posts/research/rl/revisiting_rainbow
GitHubhttps://github.com/JohanSamir/revisiting_rainbow
Paperhttps://arxiv.org/abs/2011.14826
Blog posthttps://psc-g.github.io/posts/research/rl/revisiting_rainbow
Metrics and continuity in reinforcement learningCharline Le Lan, Marc G. Bellemare, Pablo Samuel Castro AAAI, 2021https://psc-g.github.io/posts/research/rl/metrics_continuity
GitHubhttps://github.com/google-research/google-research/tree/master/rl_metrics_aaai2021
Paperhttps://arxiv.org/abs/2102.01514
Blog posthttps://psc-g.github.io/posts/research/rl/metrics_continuity
Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement LearningRishabh Agarwal, Marlos C. Machado, Pablo Samuel Castro, Marc G. Bellemare ICLR, 2021https://psc-g.github.io/posts/research/rl/pse
Paperhttps://arxiv.org/abs/2101.05265
Blog posthttps://psc-g.github.io/posts/research/rl/pse
Autonomous navigation of stratospheric balloons using reinforcement learningMarc G. Bellemare, Salvatore Candido, Pablo Samuel Castro, Jun Gong, Marlos C. Machado, Subhodeep Moitra, Sameera S. Ponda, Ziyu Wang Nature, 2020https://psc-g.github.io/posts/research/rl/loon
Paperhttps://www.nature.com/articles/s41586-020-2939-8
Blog posthttps://psc-g.github.io/posts/research/rl/loon
Estimating Policy Functions in Payment Systems using Reinforcement LearningPablo Samuel Castro, Ajit Desai, Han Du, Rodney Garratt, Francisco Rivadeneyra Best paper at ML for Economic Policy Workshop, NeurIPS 2020https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3743017
Paperhttps://papers.ssrn.com/sol3/papers.cfm?abstract_id=3743017
Dopamine, A Research Framework for Deep Reinforcement LearningPablo Samuel Castro, Subhodeep Moitra, Carles Gelada, Saurabh Kumar, and Marc G. Bellemare Preprint, 2018https://psc-g.github.io/posts/research/rl/dopamine
GitHubhttps://github.com/google/dopamine
Paperhttps://arxiv.org/abs/1812.06110
Blog posthttps://psc-g.github.io/posts/research/rl/dopamine
GANterpretationsPablo Samuel Castro Machine Learning for Creativity and Design workshop at NeurIPS, 2020https://psc-g.github.io/posts/research/creativity/ganterpretations
GitHubhttps://github.com/psc-g/ganterpretation
Blog posthttps://psc-g.github.io/posts/research/creativity/ganterpretations
Rigging the lottery, Making all tickets winnersUtku Evci, Trevor Gale, Jacob Menick, Pablo Samuel Castro, and Erich Elsen ICML, 2020https://psc-g.github.io/posts/research/other/rigl
GitHubhttps://github.com/google-research/rigl
Paperhttps://arxiv.org/abs/1911.11134
Blog posthttps://psc-g.github.io/posts/research/other/rigl
Scalable methods for computing state similarity in deterministic MDPsPablo Samuel Castro AAAI, 2020https://psc-g.github.io/posts/research/other/scalable
GitHubhttps://github.com/google-research/google-research/tree/master/bisimulation_aaai2020
Paperhttps://arxiv.org/abs/1911.09291
Blog posthttps://psc-g.github.io/posts/research/other/scalable
A Geometric Perspective on Optimal Representations for Reinforcement LearningMarc G. Bellemare, Will Dabney, Robert Dadashi, Adrien Ali Taiga, Pablo Samuel Castro, Nicolas Le Roux, Dale Schuurmans, Tor Lattimore, and Clare Lyle NeurIPS, 2019https://proceedings.neurips.cc/paper/2019/hash/3cf2559725a9fdfa602ec8c887440f32-Abstract.html
Paperhttps://proceedings.neurips.cc/paper/2019/hash/3cf2559725a9fdfa602ec8c887440f32-Abstract.html
A comparative analysis of expected and distributional reinforcement learningClare Lyle, Pablo Samuel Castro, and Marc G. Bellemare AAAI, 2019https://www.aaai.org/ojs/index.php/AAAI/article/view/4365
Paperhttps://www.aaai.org/ojs/index.php/AAAI/article/view/4365
ML-Jam, Performing Structured Improvisations with Pre-trained ModelsPablo Samuel Castro ICCC, 2019https://psc-g.github.io/posts/research/creativity/ml-jam
GitHubhttps://github.com/psc-g/Psc2
Paperhttps://arxiv.org/abs/1904.13285
Blog posthttps://psc-g.github.io/posts/research/creativity/ml-jam
Shaping the Narrative Arc, Information-Theoretic Collaborative DialogueKory Mathewson, Pablo Samuel Castro, Colin Cherry, George Foster, and Marc G. Bellemare ICCC, 2020https://arxiv.org/abs/1901.11528
Paperhttps://arxiv.org/abs/1901.11528
An Atari Model Zoo for Analyzing, Visualizing, and Comparing Deep Reinforcement Learning AgentsFelipe Petroski Such, Vashisht Madhavan, Rosanne Liu, Rui Wang, Pablo Samuel Castro, Yulun Li, Jiale Zhi, Ludwig Schubert, Marc G. Bellemare, Jeff Clune, and Joel Lehman IJCAI, 2019https://arxiv.org/abs/1812.07069
GitHubhttps://github.com/uber-research/atari-model-zoo
Paperhttps://arxiv.org/abs/1812.07069
Distributional reinforcement learning with linear function approximationMarc G. Bellemare, Nicolas Le Roux, Pablo Samuel Castro, and Subhodeep Moitra AISTATS, 2019https://arxiv.org/abs/1902.03149
Paperhttps://arxiv.org/abs/1902.03149
Inverse reinforcement learning with multiple ranked expertsPablo Samuel Castro, Shijian Li, and Daqing Zhang Preprint, 2019https://arxiv.org/abs/1907.13411
Paperhttps://arxiv.org/abs/1907.13411
Google Scholar Pagehttps://scholar.google.com/citations?user=jn5r6TsAAAAJ&hl=en
Mila directory pagehttps://mila.quebec/en/directory/pablo-samuel-castro
Website https://roger-creus.github.io/
GitHub https://github.com/roger-creus
Scholarhttps://scholar.google.com/citations?view_op=list_works&hl=en&user=E3y_txsAAAAJ
📄 ARM-FM: Automated Reward Machines via Foundation Modelshttps://openreview.net/forum?id=OBpQdCWLfd
📄 Stable Gradients for Stable Learning at Scale in Deep RLhttps://openreview.net/forum?id=Vqj65VeDOu
Website https://labchameleon.github.io
GitHub https://github.com/LabChameleon
Scholarhttps://scholar.google.com/citations?user=dYzkTPEAAAAJ&hl=en
📄 Optimistic critics can empower small actorshttps://openreview.net/forum?id=yx68Ns2e5E
Website https://waltermayor.github.io/
GitHub https://github.com/waltermayor
Scholarhttps://scholar.google.com/citations?hl=en&user=fGTJwC4AAAAJ
📄 Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agentshttps://openreview.net/forum?id=mCpq1GCKxA
📄 The Impact of On-Policy Parallelized Data Collection on Deep RL Networkshttps://openreview.net/forum?id=cnqyzuZhSo
Website https://johanobandoc.github.io
GitHub https://github.com/JohanSamir
Scholarhttps://scholar.google.com/citations?hl=es&user=KViAb3EAAAAJ
📄 Simplicial Embeddings Improve Sample Efficiency in Actor-Critic Agentshttps://openreview.net/forum?id=mCpq1GCKxA
📄 Stable Deep RL via Isotropic Gaussian Representationshttps://openreview.net/forum?id=gc7Gg18ejz
📄 Stable Gradients for Stable Learning at Scale in Deep RLhttps://openreview.net/forum?id=Vqj65VeDOu
📄 Don't flatten, tokenize! Unlocking the key to SoftMoE's efficacy in deep RLhttps://proceedings.iclr.cc/paper_files/paper/2025/hash/74d056577c164d45f713ce297a4872df-Abstract-Conference.html
📄 The Impact of On-Policy Parallelized Data Collection on Deep RL Networkshttps://openreview.net/forum?id=cnqyzuZhSo
📄 In value-based deep RL, a pruned network is a good networkhttps://arxiv.org/abs/2402.12479
📄 Mixtures of experts unlock parameter scaling for deep RLhttps://arxiv.org/abs/2402.08609
📄 On the consistency of hyper-parameter selection in value-based deep RLhttps://arxiv.org/abs/2406.17523
Website https://www.gandharvpatil.com
GitHub https://github.com/gp1702
Scholarhttps://scholar.google.com/citations?user=onD9-zcAAAAJ&hl=en
📄 Stable Deep RL via Isotropic Gaussian Representationshttps://openreview.net/forum?id=gc7Gg18ejz
Website https://dhruvsreenivas.github.io
GitHubhttps://github.com/dhruvsreenivas
📄 Optimistic critics can empower small actorshttps://openreview.net/forum?id=yx68Ns2e5E
📄 A Survey of State Representation Learning for Deep Reinforcement Learninghttps://openreview.net/forum?id=gOk34vUHtz
Website https://johanobandoc.github.io
GitHub https://github.com/JohanSamir
Scholarhttps://scholar.google.com/citations?hl=es&user=KViAb3EAAAAJ
📄 Revisiting Rainbowhttps://arxiv.org/abs/2011.14826
📄 Small batch deep reinforcement learninghttps://proceedings.neurips.cc/paper_files/paper/2023/hash/528388f1ad3a481249a97cbb698d2fe6-Abstract-Conference.html
Abouthttps://psc-g.github.io/#about
Recent Postshttps://psc-g.github.io/#recent-posts
Selected Publicationshttps://psc-g.github.io/#publications
Studentshttps://psc-g.github.io/#students
Twitter https://twitter.com/@pcastr
Mastodonhttps://sigmoid.social/@psc
Tohahttps://github.com/hossainemruz/toha
Powered by https://gohugo.io/

Viewport: width=device-width,initial-scale=1

URLs of crawlers that visited me.