# STOPS: Short-Term-based Volatility-controlled Policy Search and its Global Convergence

arXiv: https://arxiv.org/abs/2201.09857 · HTML: https://mlsanity.com/paper/2201.09857/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 35 of its 81 references that are in the map.
- First posted on arXiv 24 Jan 2022.

## Builds on (most cited first, 30 of 35)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Trust Region Policy Optimization](https://mlsanity.com/paper/1502.05477/) (2015)
- [OpenAI Gym](https://mlsanity.com/paper/1606.01540/) (2016)
- [A Convergence Theory for Deep Learning via Over-Parameterization](https://mlsanity.com/paper/1811.03962/) (2018)
- [Gradient Descent Provably Optimizes Over-parameterized Neural Networks](https://mlsanity.com/paper/1810.02054/) (2018)
- [Fine-Grained Analysis of Optimization and Generalization for Overparameterized Two-Layer Neural Networks](https://mlsanity.com/paper/1901.08584/) (2019)
- [Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers](https://mlsanity.com/paper/1811.04918/) (2018)
- [On the Global Convergence Rates of Softmax Policy Gradient Methods](https://mlsanity.com/paper/2005.06392/) (2020)
- [Coordinate Descent Algorithms](https://mlsanity.com/paper/1502.04759/) (2015)
- [Reinforcement Learning in Feature Space: Matrix Bandit, Kernels, and Regret Bound](https://mlsanity.com/paper/1905.10389/) (2019)
- [Fast Global Convergence of Natural Policy Gradient Methods with Entropy Regularization](https://mlsanity.com/paper/2007.06558/) (2020)
- [Finite-Time Error Bounds For Linear Stochastic Approximation and TD Learning](https://mlsanity.com/paper/1902.00923/) (2019)
- [Neural Policy Gradient Methods: Global Optimality and Rates of Convergence](https://mlsanity.com/paper/1909.01150/) (2019)
- [Global Optimality Guarantees For Policy Gradient Methods](https://mlsanity.com/paper/1906.01786/) (2019)
- [CRPO: A New Approach for Safe Reinforcement Learning with Convergence Guarantee](https://mlsanity.com/paper/2011.05869/) (2020)
- [Adaptive Trust Region Policy Optimization: Global Convergence and Faster Rates for Regularized MDPs](https://mlsanity.com/paper/1909.02769/) (2019)
- [Global Convergence of Langevin Dynamics Based Algorithms for Nonconvex Optimization](https://mlsanity.com/paper/1707.06618/) (2017)
- [Stochastic Compositional Gradient Descent: Algorithms for Minimizing Compositions of Expected-Value Functions](https://mlsanity.com/paper/1411.3803/) (2014)
- [Generalization Error Bounds of Gradient Descent for Learning Over-parameterized Deep ReLU Networks](https://mlsanity.com/paper/1902.01384/) (2019)
- [How Should a Robot Assess Risk? Towards an Axiomatic Theory of Risk in Robotics](https://mlsanity.com/paper/1710.11040/) (2017)
- [Risk Averse Robust Adversarial Reinforcement Learning](https://mlsanity.com/paper/1904.00511/) (2019)
- [Risk-Averse Trust Region Optimization for Reward-Volatility Reduction](https://mlsanity.com/paper/1912.03193/) (2019)
- [Single-Timescale Actor-Critic Provably Finds Globally Optimal Policy](https://mlsanity.com/paper/2008.00483/) (2020)
- [Finite-Sample Analysis of Proximal Gradient TD Algorithms](https://mlsanity.com/paper/2006.14364/) (2020)
- [Two Time-scale Off-Policy TD Learning: Non-asymptotic Analysis over Markovian Samples](https://mlsanity.com/paper/1909.11907/) (2019)
- [Mean-Variance Policy Iteration for Risk-Averse Reinforcement Learning](https://mlsanity.com/paper/2004.10888/) (2020)
- [Mean-Field Multi-Agent Reinforcement Learning: A Decentralized Network Approach](https://mlsanity.com/paper/2108.02731/) (2021)
- [Neural Temporal-Difference and Q-Learning Provably Converge to Global Optima](https://mlsanity.com/paper/1905.10027/) (2019)
- [Finite-sample Analysis of Greedy-GQ with Linear Function Approximation under Markovian Noise](https://mlsanity.com/paper/2005.10175/) (2020)
- [Global Optimality and Finite Sample Analysis of Softmax Off-Policy Actor Critic under State Distribution Mismatch](https://mlsanity.com/paper/2111.02997/) (2021)

## Abstract (arXiv)

It remains challenging to deploy existing risk-averse approaches to real-world applications. The reasons are multi-fold, including the lack of global optimality guarantee and the necessity of learning from long-term consecutive trajectories. Long-term consecutive trajectories are prone to involving visiting hazardous states, which is a major concern in the risk-averse setting. This paper proposes Short-Term VOlatility-controlled Policy Search (STOPS), a novel algorithm that solves risk-averse problems by learning from short-term trajectories instead of long-term trajectories. Short-term trajectories are more flexible to generate, and can avoid the danger of hazardous state visitations. By using an actor-critic scheme with an overparameterized two-layer neural network, our algorithm finds a globally optimal policy at a sublinear rate with proximal policy optimization and natural policy gradient, with effectiveness comparable to the state-of-the-art convergence rate of risk-neutral policy-search methods. The algorithm is evaluated on challenging Mujoco robot simulation tasks under the mean-variance evaluation metric. Both theoretical analysis and experimental results demonstrate a state-of-the-art level of STOPS' performance among existing risk-averse policy search methods.