Planning By Active Sensing

Kaushik Lakshminarasimhan, Seren Zhu, Dora Angelaki
Proceedings of The 2nd Gaze Meets ML workshop, PMLR 226:125-141, 2024.

Abstract

Flexible behavior requires rapid planning, but planning requires a good internal model of the environment. Learning this model by trial-and-error is impractical when acting in complex environments. How do humans plan action sequences efficiently when there is uncertainty about model components? To address this, we asked human participants to navigate complex mazes in virtual reality. We found that the paths taken to gather rewards were close to optimal even though participants had no prior knowledge of these environments. Based on the sequential eye movement patterns observed when participants mentally compute a path before navigating, we develop an algorithm that is capable of rapidly planning under uncertainty by active sensing i.e., visually sampling information about the structure of the environment. ew eye movements are chosen in an iterative manner by following the gradient of a dynamic value map which is updated based on the previous eye movement, until the planning process reaches convergence. In addition to bearing hallmarks of human navigational planning, the proposed algorithm is sample-efficient such that the number of visual samples needed for planning scales linearly with the path length regardless of the size of the state space.

Cite this Paper

BibTeX
@InProceedings{pmlr-v226-lakshminarasimhan24a, title = {Planning By Active Sensing}, author = {Lakshminarasimhan, Kaushik and Zhu, Seren and Angelaki, Dora}, booktitle = {Proceedings of The 2nd Gaze Meets ML workshop}, pages = {125--141}, year = {2024}, editor = {Madu Blessing, Amarachi and Wu, Joy and Zanca, Dario and Krupinski, Elizabeth and Kashyap, Satyananda and Karargyris, Alexandros}, volume = {226}, series = {Proceedings of Machine Learning Research}, month = {16 Dec}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v226/lakshminarasimhan24a/lakshminarasimhan24a.pdf}, url = {https://proceedings.mlr.press/v226/lakshminarasimhan24a.html}, abstract = {Flexible behavior requires rapid planning, but planning requires a good internal model of the environment. Learning this model by trial-and-error is impractical when acting in complex environments. How do humans plan action sequences efficiently when there is uncertainty about model components? To address this, we asked human participants to navigate complex mazes in virtual reality. We found that the paths taken to gather rewards were close to optimal even though participants had no prior knowledge of these environments. Based on the sequential eye movement patterns observed when participants mentally compute a path before navigating, we develop an algorithm that is capable of rapidly planning under uncertainty by active sensing i.e., visually sampling information about the structure of the environment. ew eye movements are chosen in an iterative manner by following the gradient of a dynamic value map which is updated based on the previous eye movement, until the planning process reaches convergence. In addition to bearing hallmarks of human navigational planning, the proposed algorithm is sample-efficient such that the number of visual samples needed for planning scales linearly with the path length regardless of the size of the state space.} }
Endnote
%0 Conference Paper %T Planning By Active Sensing %A Kaushik Lakshminarasimhan %A Seren Zhu %A Dora Angelaki %B Proceedings of The 2nd Gaze Meets ML workshop %C Proceedings of Machine Learning Research %D 2024 %E Amarachi Madu Blessing %E Joy Wu %E Dario Zanca %E Elizabeth Krupinski %E Satyananda Kashyap %E Alexandros Karargyris %F pmlr-v226-lakshminarasimhan24a %I PMLR %P 125--141 %U https://proceedings.mlr.press/v226/lakshminarasimhan24a.html %V 226 %X Flexible behavior requires rapid planning, but planning requires a good internal model of the environment. Learning this model by trial-and-error is impractical when acting in complex environments. How do humans plan action sequences efficiently when there is uncertainty about model components? To address this, we asked human participants to navigate complex mazes in virtual reality. We found that the paths taken to gather rewards were close to optimal even though participants had no prior knowledge of these environments. Based on the sequential eye movement patterns observed when participants mentally compute a path before navigating, we develop an algorithm that is capable of rapidly planning under uncertainty by active sensing i.e., visually sampling information about the structure of the environment. ew eye movements are chosen in an iterative manner by following the gradient of a dynamic value map which is updated based on the previous eye movement, until the planning process reaches convergence. In addition to bearing hallmarks of human navigational planning, the proposed algorithm is sample-efficient such that the number of visual samples needed for planning scales linearly with the path length regardless of the size of the state space.
APA
Lakshminarasimhan, K., Zhu, S. & Angelaki, D.. (2024). Planning By Active Sensing. Proceedings of The 2nd Gaze Meets ML workshop, in Proceedings of Machine Learning Research 226:125-141 Available from https://proceedings.mlr.press/v226/lakshminarasimhan24a.html.

Related Material