Tsung Yen Yang, Justinian Rosca, Karthik Narasimhan, Peter J. Ramadge

Research output: Contribution to conferencePaperpeer-review

102 Scopus citations


We consider the problem of learning control policies that optimize a reward function while satisfying constraints due to considerations of safety, fairness, or other costs. We propose a new algorithm, Projection-Based Constrained Policy Optimization (PCPO). This is an iterative method for optimizing policies in a two-step process: the first step performs a local reward improvement update, while the second step reconciles any constraint violation by projecting the policy back onto the constraint set. We theoretically analyze PCPO and provide a lower bound on reward improvement, and an upper bound on constraint violation, for each policy update. We further characterize the convergence of PCPO based on two different metrics: L2 norm and Kullback-Leibler divergence. Our empirical results over several control tasks demonstrate that PCPO achieves superior performance, averaging more than 3.5 times less constraint violation and around 15% higher reward compared to state-of-the-art methods.

Original languageEnglish (US)
StatePublished - 2020
Event8th International Conference on Learning Representations, ICLR 2020 - Addis Ababa, Ethiopia
Duration: Apr 30 2020 → …


Conference8th International Conference on Learning Representations, ICLR 2020
CityAddis Ababa
Period4/30/20 → …

All Science Journal Classification (ASJC) codes

  • Education
  • Linguistics and Language
  • Language and Linguistics
  • Computer Science Applications


Dive into the research topics of 'PROJECTION-BASED CONSTRAINED POLICY OPTIMIZATION'. Together they form a unique fingerprint.

Cite this