mdps who can raise ik you're itching to do it but if i waste my swiftcast bc you are trigger happy i'm exploding you with my mind ❤️
October 13, 2026 at 7:03 PM
mdps who can raise ik you're itching to do it but if i waste my swiftcast bc you are trigger happy i'm exploding you with my mind ❤️
S Akshay, Suguman Bansal, Gaurav Mahla: Certifying Almost-Sure Reachability in Unknown MDPs https://arxiv.org/abs/2610.18119 https://arxiv.org/pdf/2610.18119 https://arxiv.org/html/2610.18119
October 14, 2026 at 6:42 AM
S Akshay, Suguman Bansal, Gaurav Mahla: Certifying Almost-Sure Reachability in Unknown MDPs https://arxiv.org/abs/2610.18119 https://arxiv.org/pdf/2610.18119 https://arxiv.org/html/2610.18119
Gongpu Chen, Deniz G\"und\"uz: Embedding Information in Policy Randomness: Distribution-Preserving Implicit Communication in MDPs https://arxiv.org/abs/2610.15976 https://arxiv.org/pdf/2610.15976 https://arxiv.org/html/2610.15976
October 14, 2026 at 6:42 AM
Gongpu Chen, Deniz G\"und\"uz: Embedding Information in Policy Randomness: Distribution-Preserving Implicit Communication in MDPs https://arxiv.org/abs/2610.15976 https://arxiv.org/pdf/2610.15976 https://arxiv.org/html/2610.15976
Qian Zuo, Francesco Emanuele Stradi: Toward Optimal Regret in Adversarial MDPs with Stochastic Hard Constraints https://arxiv.org/abs/2610.12153 https://arxiv.org/pdf/2610.12153 https://arxiv.org/html/2610.12153
October 9, 2026 at 6:44 AM
Qian Zuo, Francesco Emanuele Stradi: Toward Optimal Regret in Adversarial MDPs with Stochastic Hard Constraints https://arxiv.org/abs/2610.12153 https://arxiv.org/pdf/2610.12153 https://arxiv.org/html/2610.12153
Mingyi Li, Taira Tsuchiya: Closing the Horizon Gap in Policy Optimization for Adversarial MDPs https://arxiv.org/abs/2610.12362 https://arxiv.org/pdf/2610.12362 https://arxiv.org/html/2610.12362
October 9, 2026 at 6:45 AM
Mingyi Li, Taira Tsuchiya: Closing the Horizon Gap in Policy Optimization for Adversarial MDPs https://arxiv.org/abs/2610.12362 https://arxiv.org/pdf/2610.12362 https://arxiv.org/html/2610.12362
very cool! In my Intro RL course this year I also avoided MDPs (at first), but with a PG focus instead in a one step setting (bsky.app/profile/sard...)
working on some lecture notes to guide my revamped undergraduate reinforcement learning course this fall. (as you can from the page numbers, its still mostly blank.) we are going to dive into policy gradients before uttering the name Markov 😲
October 7, 2026 at 12:45 PM
very cool! In my Intro RL course this year I also avoided MDPs (at first), but with a PG focus instead in a one step setting (bsky.app/profile/sard...)
🔄 Updated Arxiv Paper
Title: Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
Authors: Toshinori Kitamura, Arnob Ghosh, Alex Ayoub, Thang D. Chu, Csaba Szepesv\'ari
Read more: https://arxiv.org/abs/2604.21177
Title: Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
Authors: Toshinori Kitamura, Arnob Ghosh, Alex Ayoub, Thang D. Chu, Csaba Szepesv\'ari
Read more: https://arxiv.org/abs/2604.21177
October 9, 2026 at 8:02 AM
🔄 Updated Arxiv Paper
Title: Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
Authors: Toshinori Kitamura, Arnob Ghosh, Alex Ayoub, Thang D. Chu, Csaba Szepesv\'ari
Read more: https://arxiv.org/abs/2604.21177
Title: Revisiting Subgradient Dominance in Robust MDPs: Counterexamples, Hardness, and Sufficient Conditions
Authors: Toshinori Kitamura, Arnob Ghosh, Alex Ayoub, Thang D. Chu, Csaba Szepesv\'ari
Read more: https://arxiv.org/abs/2604.21177
Huizhen Yu, Isaiah Heidt: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach https://arxiv.org/abs/2610.10326 https://arxiv.org/pdf/2610.10326 https://arxiv.org/html/2610.10326
October 8, 2026 at 6:45 AM
Huizhen Yu, Isaiah Heidt: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach https://arxiv.org/abs/2610.10326 https://arxiv.org/pdf/2610.10326 https://arxiv.org/html/2610.10326
🔄 Updated Arxiv Paper
Title: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
Authors: Huizhen Yu, Isaiah Heidt
Read more: https://arxiv.org/abs/2610.10326
Title: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
Authors: Huizhen Yu, Isaiah Heidt
Read more: https://arxiv.org/abs/2610.10326
October 8, 2026 at 8:02 AM
🔄 Updated Arxiv Paper
Title: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
Authors: Huizhen Yu, Isaiah Heidt
Read more: https://arxiv.org/abs/2610.10326
Title: Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
Authors: Huizhen Yu, Isaiah Heidt
Read more: https://arxiv.org/abs/2610.10326
Zihao Zhao, Ashwath K. Karunakaram, Ali Eshragh, Yuexing Li, Kai Wang: Decision-Focused Learning in MDPs: An Occupancy Measure Approach https://arxiv.org/abs/2610.08384 https://arxiv.org/pdf/2610.08384 https://arxiv.org/html/2610.08384
October 7, 2026 at 6:45 AM
Zihao Zhao, Ashwath K. Karunakaram, Ali Eshragh, Yuexing Li, Kai Wang: Decision-Focused Learning in MDPs: An Occupancy Measure Approach https://arxiv.org/abs/2610.08384 https://arxiv.org/pdf/2610.08384 https://arxiv.org/html/2610.08384
Salah Chikhi, Abdelhaq Chaoui, Asuman Ozdaglar, Saurabh Amin: Taylor Representations for Model-Free RL in Networked MDPs https://arxiv.org/abs/2610.05456 https://arxiv.org/pdf/2610.05456 https://arxiv.org/html/2610.05456
October 6, 2026 at 7:03 AM
Salah Chikhi, Abdelhaq Chaoui, Asuman Ozdaglar, Saurabh Amin: Taylor Representations for Model-Free RL in Networked MDPs https://arxiv.org/abs/2610.05456 https://arxiv.org/pdf/2610.05456 https://arxiv.org/html/2610.05456
Enes Arda, Atilla Eryilmaz: Sharp Convergence and Sample Complexity of Policy Mirror Descent for Average-Reward MDPs https://arxiv.org/abs/2610.04117 https://arxiv.org/pdf/2610.04117 https://arxiv.org/html/2610.04117
October 6, 2026 at 6:42 AM
Enes Arda, Atilla Eryilmaz: Sharp Convergence and Sample Complexity of Policy Mirror Descent for Average-Reward MDPs https://arxiv.org/abs/2610.04117 https://arxiv.org/pdf/2610.04117 https://arxiv.org/html/2610.04117
I think it was because of the MDPs response that Platner slow-walked his resignation filing with the Maine Dems, delaying their preparations for a renomination process by almost a week after he announced he would resign. It was his and Katz's final "fuck you" to Maine Democrats.
October 5, 2026 at 5:34 PM
I think it was because of the MDPs response that Platner slow-walked his resignation filing with the Maine Dems, delaying their preparations for a renomination process by almost a week after he announced he would resign. It was his and Katz's final "fuck you" to Maine Democrats.
New work on safe RL tackles step-wise constrained MDPs with Safe Variance-Adaptive Exploration, achieving instance-dependent regret that adapts to the variance structure of feasible policies. The result highlights how characterizing safe…
#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.02520
#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.02520
October 6, 2026 at 4:01 AM
New work on safe RL tackles step-wise constrained MDPs with Safe Variance-Adaptive Exploration, achieving instance-dependent regret that adapts to the variance structure of feasible policies. The result highlights how characterizing safe…
#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.02520
#AI #AIResearch #TechNews
https://arxiv.org/abs/2610.02520
A study reveals that success conditioning converges to optimal policies in Markov decision processes, demonstrating convergence rates for discounted and single-period MDPs. This stochastic approach may transform AI training for uncertain decision-making. https://arxiv.org/abs/2610.03642
On the Convergence of Success Conditioning for Policy Optimization
ArXiv link for On the Convergence of Success Conditioning for Policy Optimization
arxiv.org
October 5, 2026 at 10:30 PM
A study reveals that success conditioning converges to optimal policies in Markov decision processes, demonstrating convergence rates for discounted and single-period MDPs. This stochastic approach may transform AI training for uncertain decision-making. https://arxiv.org/abs/2610.03642
Q-Learning for Reachability in MEC-Free MDPs
Lu-Chin Chang, Suguman Bansal
#arXiv #cs.AI #cs.LG #cs.LO
Lu-Chin Chang, Suguman Bansal
#arXiv #cs.AI #cs.LG #cs.LO
Q-Learning for Reachability in MEC-Free MDPs
Reinforcement learning (RL) for reachability specifications is fundamental to sequential decision-making. Prior work establishes asymptotic convergence to optimal policies, but only through model-based methods that must explicitly estimate the transition probabilities of the underlying Markov Decis…
arxiv.org
October 2, 2026 at 9:45 PM
Lu-Chin Chang, Suguman Bansal: Q-Learning for Reachability in MEC-Free MDPs https://arxiv.org/abs/2610.01781 https://arxiv.org/pdf/2610.01781 https://arxiv.org/html/2610.01781
October 2, 2026 at 6:40 AM
Lu-Chin Chang, Suguman Bansal: Q-Learning for Reachability in MEC-Free MDPs https://arxiv.org/abs/2610.01781 https://arxiv.org/pdf/2610.01781 https://arxiv.org/html/2610.01781
Part of I-55, other roads to close for old MDPS tower demolition
Part of I-55, other roads to close for old MDPS tower demolition
JACKSON, Miss. (WJTV) – The former Mississippi Department of Public Safety (MDPS) administration tower off Woodrow Wilson Avenue in Jackson has been scheduled for demolition on October 4, 202…
www.wjtv.com
October 1, 2026 at 7:01 PM
Part of I-55, other roads to close for old MDPS tower demolition
Zhaojun Peng: Bellman-Certified Rounding for Sparse Policy Deployment in MDPs https://arxiv.org/abs/2610.00325 https://arxiv.org/pdf/2610.00325 https://arxiv.org/html/2610.00325
October 2, 2026 at 6:42 AM
Zhaojun Peng: Bellman-Certified Rounding for Sparse Policy Deployment in MDPs https://arxiv.org/abs/2610.00325 https://arxiv.org/pdf/2610.00325 https://arxiv.org/html/2610.00325
A study finds Howard's policy iteration for deterministic MDPs is not strongly polynomial, demonstrating an exponential iteration lower bound. This reveals the contrast with Dantzig's simplex method, illustrating inefficiencies in decentralized algorithms. https://arxiv.org/abs/2609.40147
Policy Iteration Is Not Strongly Polynomial for Deterministic Markov Decision Processes: The Price of Algorithmic Anarchy
ArXiv link for Policy Iteration Is Not Strongly Polynomial for Deterministic Markov Decision Processes: The Price of Algorithmic Anarchy
arxiv.org
October 2, 2026 at 12:50 PM
A study finds Howard's policy iteration for deterministic MDPs is not strongly polynomial, demonstrating an exponential iteration lower bound. This reveals the contrast with Dantzig's simplex method, illustrating inefficiencies in decentralized algorithms. https://arxiv.org/abs/2609.40147
No matter what you do in a gacha game, if you main dps a male character who is currently meta, always know that a cooler and better mdps female character will drop in a month
September 29, 2026 at 7:17 AM
No matter what you do in a gacha game, if you main dps a male character who is currently meta, always know that a cooler and better mdps female character will drop in a month
🔄 Updated Arxiv Paper
Title: Achieving an $O(1/N)$ Optimality Gap in Average-Reward Weakly-Coupled MDPs
Authors: Yige Hong, Xiangcheng Zhang, Qiaomin Xie, Yudong Chen, Weina Wang
Read more: https://arxiv.org/abs/2609.38132
Title: Achieving an $O(1/N)$ Optimality Gap in Average-Reward Weakly-Coupled MDPs
Authors: Yige Hong, Xiangcheng Zhang, Qiaomin Xie, Yudong Chen, Weina Wang
Read more: https://arxiv.org/abs/2609.38132
September 30, 2026 at 8:02 AM
🔄 Updated Arxiv Paper
Title: Achieving an $O(1/N)$ Optimality Gap in Average-Reward Weakly-Coupled MDPs
Authors: Yige Hong, Xiangcheng Zhang, Qiaomin Xie, Yudong Chen, Weina Wang
Read more: https://arxiv.org/abs/2609.38132
Title: Achieving an $O(1/N)$ Optimality Gap in Average-Reward Weakly-Coupled MDPs
Authors: Yige Hong, Xiangcheng Zhang, Qiaomin Xie, Yudong Chen, Weina Wang
Read more: https://arxiv.org/abs/2609.38132
Nam Phuong Tran, Trinh Ha Mai Huynh, Tuyen Pham Le, Van-Truong Nguyen, Quan Nguyen, Long Tran-Thanh: Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs https://arxiv.org/abs/2609.32933 https://arxiv.org/pdf/2609.32933 https://arxiv.org/html/2609.32933
September 30, 2026 at 6:45 AM
Nam Phuong Tran, Trinh Ha Mai Huynh, Tuyen Pham Le, Van-Truong Nguyen, Quan Nguyen, Long Tran-Thanh: Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs https://arxiv.org/abs/2609.32933 https://arxiv.org/pdf/2609.32933 https://arxiv.org/html/2609.32933
I posted on the GI subreddit about Eula, Freminet and Heizou getting robbed of a Radiance buff, but Lynette is truly the biggest missed opportunity. Her whole kit is literally perfect for it, down to the unexplained CPHE priority, AND her C6 making her playable as MDPS. The ideal SSw Anemo Sub-DPS.
September 29, 2026 at 8:30 PM
I posted on the GI subreddit about Eula, Freminet and Heizou getting robbed of a Radiance buff, but Lynette is truly the biggest missed opportunity. Her whole kit is literally perfect for it, down to the unexplained CPHE priority, AND her C6 making her playable as MDPS. The ideal SSw Anemo Sub-DPS.
Nam Phuong Tran, Trinh Ha Mai Huynh, Tuyen Pham Le, Van-Truong Nguyen, Quan Nguyen, Long Tran-Thanh: Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs https://arxiv.org/abs/2609.32933 https://arxiv.org/pdf/2609.32933 https://arxiv.org/html/2609.32933
September 29, 2026 at 6:45 AM
Nam Phuong Tran, Trinh Ha Mai Huynh, Tuyen Pham Le, Van-Truong Nguyen, Quan Nguyen, Long Tran-Thanh: Last-Iterate Guarantees for Online Reinforcement Learning in Structured Constrained MDPs https://arxiv.org/abs/2609.32933 https://arxiv.org/pdf/2609.32933 https://arxiv.org/html/2609.32933