No article text was recorded for this version.
softmax gradient vanishing (policy saturation)
Version 1 of 142 papers59 conceptsfrozen 2026-10-04
Papers in this version
- Blog: Does Stochastic Gradient really succeed for bandits?
- Blog: Softmax Policy Gradient Methods Can Take Exponential Time to Converge
- Blog: On the Global Convergence Rates of Decentralized Softmax Gradient Play in Markov Potential Games
- Blog: Softmax policy gradient methods can take exponential time to converge
- Blog: Vanishing Gradients in Reinforcement Finetuning of Language Models
- Blog: The Value Function Polytope in Reinforcement Learning
- Blog: The Sufficiency of Off-Policyness and Soft Clipping: PPO Is Still Insufficient according to an Off-Policy Measure
- Blog: Beyond the Policy Gradient Theorem for Efficient Policy Updates in Actor-Critic Algorithms
- Blog: Benchmarking Actor-Critic Deep Reinforcement Learning Algorithms for Robotics Control with Action Constraints
- Blog: Beyond Softmax and Entropy: Convergence Rates of Policy Gradients with $\boldsymbol{f}$-SoftArgmax Parameterization $\&$ Coupled Regularization
- Blog: An Alternate Policy Gradient Estimator for Softmax Policies
- Blog: Surrogate Objectives for Batch Policy Optimization in One-step Decision Making
- Blog: Optimistic policy iteration and natural actor-critic: A unifying view and a non-optimality result
- Blog: Target Entropy Annealing for Discrete Soft Actor-Critic
- Blog: Independent Natural Policy Gradient always converges in Markov Potential Games
- Blog: Policy Learning for Continuous Space Security Games Using Neural Networks
- Blog: Deep Reinforcement Learning in Parameterized Action Space [data] Matthew Hausknecht
- Blog: Applying the Episodic Natural Actor-Critic Architecture to Motor Primitive Learning
- Blog: Exponential Smoothing for Off-Policy Learning
- Blog: Fast Global Convergence of Natural Policy Gradient Methods with Entropy Regularization
- Blog: Towards Principled, Practical Policy Gradient for Bandits and Tabular MDPs
- Blog: Return Capping: Sample Efficient CVaR Policy Gradient Optimisation
- Blog: Beyond Softmax and Entropy: Improving Convergence Guarantees of Policy Gradients by f-SoftArgmax Parameterization with Coupled Regularization
- Blog: On the Theory of Policy Gradient Methods: Optimality, Approximation, and Distribution Shift
- Blog: Is High Variance Unavoidable in RL? A Case Study in Continuous Control
- Blog: Optimizing Solution-Samplers for Combinatorial Problems: The Landscape of Policy-Gradient Method
- Blog: Regularized Contextual Bandits
- Blog: Neural Frank-Wolfe Policy Optimization for Region-of-Interest Intra-Frame Coding with HEVC/H.265
- Blog: The Role of Baselines in Policy Gradient Optimization
- Blog: Accelerated Policy Gradient: On the Nesterov Momentum for Reinforcement Learning
- Blog: Simple Policy Optimization
- Blog: On the Global Convergence Rates of Softmax Policy Gradient Methods
- Blog: Vanishing Gradients in Reinforcement Finetuning of Language Models
- Blog: Softmax Deep Double Deterministic Policy Gradients
- Blog: A Simple Mixture Policy Parameterization for Improving Sample Efficiency of CVaR Optimization
- Blog: FlowPG: Action-constrained Policy Gradient with Normalizing Flows
- Blog: Complexity-Aware Deep Symbolic Regression with Robust Risk-Seeking Policy Gradients
- Blog: Mollification Effects of Policy Gradient Methods
- Blog: Policy Optimization Algorithm with Activation Likelihood-Ratio for Multi-agent Reinforcement Learning
- Blog: Learning Policies from Self-Play with Policy Gradients and MCTS Value Estimates
- Blog: Adaptive Step-size Policy Gradients with Average Reward Metric
- Blog: Escaping the Gravitational Pull of Softmax
History & citation
Current · Permanent link
- Optimization Geometry of Softmax Policy Gradient Methods merged into this direction (53% overlap)
Cite this direction
Latest version (a living page):
@misc{lacuna_softmax_gradient_vanishing_policy_satura_b9c6c8ea,
title = {softmax gradient vanishing (policy saturation)},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/softmax-gradient-vanishing-policy-saturation/rdr_b9c6c8eabe974b27973110904ed96402},
note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}This exact version (version 1, frozen text):
@misc{lacuna_softmax_gradient_vanishing_policy_satura_b9c6c8ea_v1,
title = {softmax gradient vanishing (policy saturation)},
author = {{Lacuna}},
howpublished = {Lacuna research direction},
year = {2026},
url = {https://lacuna.tiptreesystems.com/direction/softmax-gradient-vanishing-policy-saturation/rdr_b9c6c8eabe974b27973110904ed96402/v/1},
note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}