Learning and planning with logical automata
Abstract We introduce a method to learn policies from expert demonstrations that are interpretable and manipulable. We achieve interpretability by modeling the interactions between high-level actions as an automaton with connections to formal logic. We achieve manipulability by integr...
| Main Authors: | , , , , , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer US
2021
|
| Online Access: | https://hdl.handle.net/1721.1/138132 |
| _version_ | 1826218095175270400 |
|---|---|
| author | Araki, Brandon Vodrahalli, Kiran Leech, Thomas Vasile, Cristian-Ioan Donahue, Mark Rus, Daniela |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Araki, Brandon Vodrahalli, Kiran Leech, Thomas Vasile, Cristian-Ioan Donahue, Mark Rus, Daniela |
| author_sort | Araki, Brandon |
| collection | MIT |
| description | Abstract
We introduce a method to learn policies from expert demonstrations that are interpretable and manipulable. We achieve interpretability by modeling the interactions between high-level actions as an automaton with connections to formal logic. We achieve manipulability by integrating this automaton into planning via Logical Value Iteration, so that changes to the automaton have predictable effects on the learned behavior. These qualities allow a human user to first understand what the model has learned, and then either correct the learned behavior or zero-shot generalize to new, similar tasks. Our inference method requires only low-level trajectories and a description of the environment in order to learn high-level rules. We achieve this by using a deep Bayesian nonparametric hierarchical model. We test our model on several domains of interest and also show results for a real-world implementation on a mobile robotic arm platform for lunchbox-packing and cabinet-opening tasks. |
| first_indexed | 2024-09-23T17:13:48Z |
| format | Article |
| id | mit-1721.1/138132 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T17:13:48Z |
| publishDate | 2021 |
| publisher | Springer US |
| record_format | dspace |
| spelling | mit-1721.1/1381322023-04-18T18:31:23Z Learning and planning with logical automata Araki, Brandon Vodrahalli, Kiran Leech, Thomas Vasile, Cristian-Ioan Donahue, Mark Rus, Daniela Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Lincoln Laboratory Abstract We introduce a method to learn policies from expert demonstrations that are interpretable and manipulable. We achieve interpretability by modeling the interactions between high-level actions as an automaton with connections to formal logic. We achieve manipulability by integrating this automaton into planning via Logical Value Iteration, so that changes to the automaton have predictable effects on the learned behavior. These qualities allow a human user to first understand what the model has learned, and then either correct the learned behavior or zero-shot generalize to new, similar tasks. Our inference method requires only low-level trajectories and a description of the environment in order to learn high-level rules. We achieve this by using a deep Bayesian nonparametric hierarchical model. We test our model on several domains of interest and also show results for a real-world implementation on a mobile robotic arm platform for lunchbox-packing and cabinet-opening tasks. 2021-11-15T13:15:37Z 2021-11-15T13:15:37Z 2021-08-13 2021-11-14T04:12:58Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/138132 Araki, Brandon, Vodrahalli, Kiran, Leech, Thomas, Vasile, Cristian-Ioan, Donahue, Mark et al. 2021. "Learning and planning with logical automata." PUBLISHER_CC en https://doi.org/10.1007/s10514-021-09993-6 Creative Commons Attribution https://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer US Springer US |
| spellingShingle | Araki, Brandon Vodrahalli, Kiran Leech, Thomas Vasile, Cristian-Ioan Donahue, Mark Rus, Daniela Learning and planning with logical automata |
| title | Learning and planning with logical automata |
| title_full | Learning and planning with logical automata |
| title_fullStr | Learning and planning with logical automata |
| title_full_unstemmed | Learning and planning with logical automata |
| title_short | Learning and planning with logical automata |
| title_sort | learning and planning with logical automata |
| url | https://hdl.handle.net/1721.1/138132 |
| work_keys_str_mv | AT arakibrandon learningandplanningwithlogicalautomata AT vodrahallikiran learningandplanningwithlogicalautomata AT leechthomas learningandplanningwithlogicalautomata AT vasilecristianioan learningandplanningwithlogicalautomata AT donahuemark learningandplanningwithlogicalautomata AT rusdaniela learningandplanningwithlogicalautomata |