Zobrazeno 1 - 10
of 20
pro vyhledávání: '"Rohanimanesh, Khashayar"'
Autor:
Booher, Jonathan, Rohanimanesh, Khashayar, Xu, Junhong, Isenbaev, Vladislav, Balakrishna, Ashwin, Gupta, Ishan, Liu, Wei, Petiushko, Aleksandr
Modern approaches to autonomous driving rely heavily on learned components trained with large amounts of human driving data via imitation learning. However, these methods require large amounts of expensive data collection and even then face challenge
Externí odkaz:
http://arxiv.org/abs/2406.08878
Autor:
Zhou, Zihan, Booher, Jonathan, Rohanimanesh, Khashayar, Liu, Wei, Petiushko, Aleksandr, Garg, Animesh
Safe reinforcement learning tasks are a challenging domain despite being very common in the real world. The widely adopted CMDP model constrains the risks in expectation, which makes room for dangerous behaviors in long-tail states. In safety-critica
Externí odkaz:
http://arxiv.org/abs/2402.15650
Deep learning-based grasp prediction models have become an industry standard for robotic bin-picking systems. To maximize pick success, production environments are often equipped with several end-effector tools that can be swapped on-the-fly, based o
Externí odkaz:
http://arxiv.org/abs/2302.07940
Robots have the capability to collect large amounts of data autonomously by interacting with objects in the world. However, it is often not obvious \emph{how} to learning from autonomously collected data without human-labeled supervision. In this wor
Externí odkaz:
http://arxiv.org/abs/2008.11466
We investigate a model for planning under uncertainty with temporallyextended actions, where multiple actions can be taken concurrently at each decision epoch. Our model is based on the options framework, and combines it with factored state space mod
Externí odkaz:
http://arxiv.org/abs/1301.2307
Autor:
Rohanimanesh, Khashayar
Publikováno v:
Doctoral Dissertations Available from Proquest.
This dissertation investigates concurrent decision making and coordination in systems that can simultaneously execute multiple actions to perform tasks more efficiently. Concurrent decision-making is a fundamental problem in many areas of robotics, c
Externí odkaz:
https://scholarworks.umass.edu/dissertations/AAI3206209
Publikováno v:
MIT web domain
URL to paper listed on conference page
Choosing features for the critic in actor-critic algorithms with function approximation is known to be a challenge. Too few critic features can lead to degeneracy of the actor gradient, and too many feature
Choosing features for the critic in actor-critic algorithms with function approximation is known to be a challenge. Too few critic features can lead to degeneracy of the actor gradient, and too many feature
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od________88::b5234ba1e93c5248271f9bebf1f6aadc
http://hdl.handle.net/1721.1/64445
http://hdl.handle.net/1721.1/64445
Publikováno v:
McCallum, A, Rohanimanesh, K & Sutton, C 2003, Dynamic Conditional Random Fields for Jointly Labeling Multiple Sequences . in NIPS Workshop on Syntax, Semantics, and Statistics .
Conditional random fields (CRFs) for sequence modeling have several advantages over joint models such as HMMs, including the ability to relax strong independence assumptions made in those models, and the ability to incorporate arbitrary overlapping f
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=od______3094::07660e90871dc989d997a9db1bcc3709
https://hdl.handle.net/20.500.11820/0cd74788-66dd-4180-8166-eec0aecb97cf
https://hdl.handle.net/20.500.11820/0cd74788-66dd-4180-8166-eec0aecb97cf
Publikováno v:
Proceeding of the 14th ACM SIGKDD International Conference: Knowledge Discovery & Data Mining; 8/24/2008, p722-730, 9p
Publikováno v:
ACM International Conference Proceeding Series; Aug2005, p720-727, 8p