Reinforcement Learning by Probability Matching

By Long Beach, Ulrike von Luxburg, Samy Bengio, Rob Fergus, Roman Garnett, I. Guyon, H. Wallach, S.V.N. Vishwanathan. 1995

Citations in the corpus

• Delayed Impact of Fair Machine Learning (2018, neutral): “These trade-offs carry over to some extent to the case where we only equalize true positive rates [Pleiss et al., 2017].”
• The Effect of Planning Shape on Dyna-style Planning in High-dimensional State Spaces (2018, neutral): “[Tamar et al., 2016; Oh et al., 2017; Weber et al., 2017]), these methods arguably discard many of the anticipated benefits of model-learning by making the model essentially a part of the parameterization of the value function or policy.”
• Can neural networks acquire a structural bias from raw linguistic data? (2020, unknown): “The Transformer neural network architecture (Vaswani et al., 2017) used in models like BERT has very weak biases: It is a universal approximator of the class of sequence transduction functions (Yun, Bhojanapalli, Rawat, Reddi, & Kumar, 2019), and it has been applied effectively in nonlinguistic…”