Compact Spectral Bases for Value Function Approximation Using Kronecker Factorization

By Jeffrey Johns, Sridhar Mahadevan, Chang Wang. 2007

16 citations recorded by the source corpus.

Subfields: Reinforcement Learning, Numerical Methods & Scientific Computing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Reinforcement Learning

1. Sergey Levine
2. Pieter Abbeel
3. George Dimitri Konidaris
4. Chelsea Finn
5. Ofir Nachum
6. Shixiang Shane Gu
7. Shimon Whiteson
8. Lihong Li

In Numerical Methods & Scientific Computing

1. Mauro Maggioni
2. Pieter Abbeel
3. Sergey Levine
4. Sridhar Mahadevan

Citations in the corpus

• Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor (2018, background): “, 2015) or Kronecker factorization (Johns et al., 2007)) or more expressive policy classes is an exciting avenue for future work.”
• Proto-value Functions: A Laplacian Framework for Learning Representation and Control in Markov Decision Processes (2007, methodology): “Section 9 discusses several ongoing extensions of the proposed framework, including Kronecker product matrix factorization (Johns et al., 2007), multiscale diffusion wavelet bases (Mahadevan and Maggioni, 2006), and more elaborate diffusion models using directed graphs where actions are part of the…”
• Value Function Approximation in Reinforcement Learning Using the Fourier Basis (2011, background): “Johns et al. (2007); Mahadevan and Maggioni (2007)) was beyond the scope of this work.”