Modeling Human Understanding of Complex Intentional Action with a Bayesian Nonparametric Subgoal Model

By Ryo Nakahashi, Chris L. Baker, Josh Tenenbaum. 2015

13 citations recorded by the source corpus.

Most human behaviors consist of multiple parts, steps, or subtasks. These structures guide our ac- tion planning and execution, but when we observe others, the latent structure of their actions is typ- ically unobservable, and must be inferred in order to learn new skills by demonstration, or to as- sist others in completing their tasks. For example, an assistant who has learned the subgoal struc- ture of a colleague’s task can more rapidly rec- ognize and support their actions as they unfold. Here we model how humans infer subgoals from observations of complex action sequences using a nonparametric Bayesian model, which assumes that observed actions are generated by approxi- mately rational planning over unknown subgoal sequences. We test this model with a behavioral experiment in which humans observed diﬀerent se- ries of goal-directed actions, and inferred both the number and composition of the subgoal sequences associated with each goal. The Bayesian model predicts human subgoal inferences with high ac- curacy, and signiﬁcantly better than several al- ternative models and straightforward heuristics. Motivated by this result, we simulate how learn- ing and inference of subgoals can improve perfor- mance in an artiﬁcial user assistance task. The Bayesian model learns the correct subgoals from fewer observations, and better assists users by more rapidly and accurately inferring the goal of their actions than alternative approaches.

Subfields: Artificial Intelligence, Developmental and Educational Psychology

How researchers see this paper

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

In Artificial Intelligence

In Developmental and Educational Psychology

Citations in the corpus

Machine Theory of Mind (2018, methodology): “…learning (e.g. Ng et al., 2000; Ramachandran & Amir, 2007; Ziebart et al., 2008; Boularias et al., 2011), or Bayesian ToM (e.g. Baker et al., 2011; Nakahashi et al., 2016; Baker et al., 2017), we drive the ToMnet to amortise its inference through neural networks (as in Kingma & Welling, 2013;…”
Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense (2020, background): “In addition, Bayesian networks, with their flexibility in representing probabilistic dependencies and causal relationships, as well as the efficiency of inference methods, have proven to be one of the most powerful and successful approaches for intent recognition [324, 325, 326, 323].”