The Task Specification Problem

Pulkit Agrawal
Proceedings of the 5th Conference on Robot Learning, PMLR 164:1745-1751, 2022.

Abstract

Robots are commonly used for several industrial applications and some have made their mark even in households (e.g., the roomba). Undoubtedly these systems are impressive! However, they are very narrow in their functionality and we are not even close to building a robot butler. A central challenge is the ability to work with sensory observations and generalization to novel situations. While we do not prescribe a solution to this problem, we do provide a perspective on a few dominant ideas in robot learning for multi-task learning and generalization. This perspective suggests a counter-intuitive conclusion: the primary challenge in building generalizable robotic systems (e.g., a robot butler) is not in the learning algorithms or the hardware, but in how humans transfer their knowledge to robots.

Cite this Paper


BibTeX
@InProceedings{pmlr-v164-agrawal22a, title = {The Task Specification Problem }, author = {Agrawal, Pulkit}, booktitle = {Proceedings of the 5th Conference on Robot Learning}, pages = {1745--1751}, year = {2022}, editor = {Faust, Aleksandra and Hsu, David and Neumann, Gerhard}, volume = {164}, series = {Proceedings of Machine Learning Research}, month = {08--11 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v164/agrawal22a/agrawal22a.pdf}, url = {https://proceedings.mlr.press/v164/agrawal22a.html}, abstract = {Robots are commonly used for several industrial applications and some have made their mark even in households (e.g., the roomba). Undoubtedly these systems are impressive! However, they are very narrow in their functionality and we are not even close to building a robot butler. A central challenge is the ability to work with sensory observations and generalization to novel situations. While we do not prescribe a solution to this problem, we do provide a perspective on a few dominant ideas in robot learning for multi-task learning and generalization. This perspective suggests a counter-intuitive conclusion: the primary challenge in building generalizable robotic systems (e.g., a robot butler) is not in the learning algorithms or the hardware, but in how humans transfer their knowledge to robots. } }
Endnote
%0 Conference Paper %T The Task Specification Problem %A Pulkit Agrawal %B Proceedings of the 5th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2022 %E Aleksandra Faust %E David Hsu %E Gerhard Neumann %F pmlr-v164-agrawal22a %I PMLR %P 1745--1751 %U https://proceedings.mlr.press/v164/agrawal22a.html %V 164 %X Robots are commonly used for several industrial applications and some have made their mark even in households (e.g., the roomba). Undoubtedly these systems are impressive! However, they are very narrow in their functionality and we are not even close to building a robot butler. A central challenge is the ability to work with sensory observations and generalization to novel situations. While we do not prescribe a solution to this problem, we do provide a perspective on a few dominant ideas in robot learning for multi-task learning and generalization. This perspective suggests a counter-intuitive conclusion: the primary challenge in building generalizable robotic systems (e.g., a robot butler) is not in the learning algorithms or the hardware, but in how humans transfer their knowledge to robots.
APA
Agrawal, P.. (2022). The Task Specification Problem . Proceedings of the 5th Conference on Robot Learning, in Proceedings of Machine Learning Research 164:1745-1751 Available from https://proceedings.mlr.press/v164/agrawal22a.html.

Related Material