Bayesian inference of temporal task specifications from demonstrations
Author(s)Shah, Ankit Jayesh; Kamath, Pritish; Li, Shen; Shah, Julie A
MetadataShow full item record
When observing task demonstrations, human apprentices are able to identify whether a given task is executed correctly long before they gain expertise in actually performing that task. Prior research into learning from demonstrations (LfD) has failed to capture this notion of the acceptability of an execution; meanwhile, temporal logics provide a flexible language for expressing task specifications. Inspired by this, we present Bayesian specification inference, a probabilistic model for inferring task specification as a temporal logic formula. We incorporate methods from probabilistic programming to define our priors, along with a domain-independent likelihood function to enable sampling-based inference. We demonstrate the efficacy of our model for inferring specifications with over 90% similarity between the inferred specification and the ground truth, both within a synthetic domain and a real-world table setting task.
Paper presented at the Annual Conference on Neural Information Processing Systems 2018 (NeurIPS 2018), 3-8 December 3-8, 2018, Montréal, Québec.
DepartmentMassachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Advances in Neural Information Processing Systems
Neural Information Processing Systems Foundation, Inc.
Shah, Ankit, "Bayesian inference of temporal task specifications from demonstrations." Advances in Neural Information Processing Systems 31 (NIPS 2018), edited by S. Bengio, et al. (San Diego, Calif.: Neural Information Processing Systems Foundation, 2018): url https://papers.nips.cc/paper/7637-bayesian-inference-of-temporal-task-specifications-from-demonstrations
Final published version