Goal Inference as Inverse Planning

By Chris L. Baker, Josh Tenenbaum, Rebecca Saxe. 2007

187 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/72a36acfdd23ec56821a7756929639d0dc161f23

How researchers see this paper

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

Citations in the corpus

Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense (2020, background): “[104] proposed a framework for goal inference, in which the bottom-up information of behavior observations and the top-down prior knowledge of goal space are integrated to allow inference of underlying intent.”
Quantilizers: A Safer Alternative to Maximizers for Limited Optimization (2016, background): “(See Ng and Russell [17] and Baker, Tenenbaum, and Saxe [18] for some initial work in this direction.)”
Bayesian models of cognition (2008, unknown): “…acquisition (Chater & Manning, 2006; Xu & Tenenbaum, in press), symbolic reasoning (Oaksford & Chater, 2001), causal learning and inference (Steyvers, Tenenbaum, Wagenmakers, & Blum, 2003; Griﬃths & Tenenbaum, 2005, 2007a), and social cognition (Baker, Tenenbaum, & Saxe, 2007), among other topics.”
Feature Construction for Inverse Reinforcement Learning (2010, background): “In fact, models that perform a variant of IRL have been shown to correspond well to goal inference in humans [2].”
An Alternative Softmax Operator for Reinforcement Learning (2016, background): “The issue arises in the context of the exploration–exploitation dilemma (Thrun, 1992), non-stationary decision problems (Sutton, 1990), and when interpreting observed decisions (Baker et al., 2007).”
Learning Robot Objectives from Physical Human Interaction (2017, background): “Our choice of (2) stems from maximum entropy assumptions [8], as well as the Bolzmann distributions used in cognitive science models of human behavior [16].”
Learning a Prior over Intent via Meta-Inverse Reinforcement Learning (2018, background): “, reward functions) of even abstract agents from just one or a few demonstrations (Baker et al., 2007), as humans have access to strong priors about how other humans accomplish similar tasks accrued over many years.”
Inverse Optimal Control with Linearly-Solvable MDPs (2010, background): “The former is relevant to cognitive and neural science, where researches are interested in ”theories of mind” (Baker et al., 2007) as well as in identifying the cost functions being optimized by the sensorimotor system (Todorov, 2004; Körding & Wolpert, 2004).”
A New Softmax Operator for Reinforcement Learning (2016, background): “The issue arises in the context of the exploration–exploitation dilemma (Thrun, 1992), non-stationary decision problems (Sutton, 1990), and when interpreting observed decisions (Baker et al., 2007).”
Feature Expansive Reward Learning: Rethinking Human Input (2020, background): “where the human chooses trajectories proportional to their exponentiated reward [17, 18].”