Zobrazeno 1 - 10
of 12
pro vyhledávání: '"Varun Raj Kompella"'
Publikováno v:
Frontiers in Neurorobotics, Vol 7 (2013)
Curiosity Driven Modular Incremental Slow Feature Analysis (CD-MISFA;~cite{cdmisfa}) is a recently introduced model of intrinsically-motivated invariance learning, which shows how curiosity enables the orderly formation of multiple stable sensory rep
Externí odkaz:
https://doaj.org/article/c9a8708011ac42e3ac7f756f0f84102c
Autor:
Stacy Jong, Guni Sharon, Peter Stone, Roberto Capobianco, Spencer J. Fox, Peter R. Wurman, Varun Raj Kompella, Lauren Ancel Meyers, James Ault
Publikováno v:
Journal of Artificial Intelligence Research. 71:953-992
The year 2020 saw the covid-19 virus lead to one of the worst global pandemics in history. As a result, governments around the world have been faced with the challenge of protecting public health while keeping the economy running to the greatest exte
Autor:
Franziska Eckert, James MacGlashan, Kenta Kawamoto, Michael Spranger, Hiroaki Kitano, Varun Raj Kompella, Michael Thomure, Craig Sherstan, Leilani Gilpin, Rory Douglas, Takuma Seno, Peter R. Wurman, Florian Fuchs, Peter Stone, Roberto Capobianco, Houmehr Aghabozorgi, Alisa Devlic, Peter Durr, Dion Whitehead, Samuel Barrett, Leon Barrett, Patrick MacAlpine, HaoChih Lin, Piyush Khandelwal, Kaushik Subramanian, Declan Oller, Tom Walsh
The authors have requested that this preprint be removed from Research Square.
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::fe59ead6e8492a51c892c659cb391aa5
https://doi.org/10.21203/rs.3.rs-795954/v1
https://doi.org/10.21203/rs.3.rs-795954/v1
Publikováno v:
Artificial Intelligence. 247:313-335
In the absence of external guidance, how can a robot learn to map the many raw pixels of high-dimensional visual inputs to useful action sequences? We propose here Continual Curiosity driven Skill Acquisition (CCSA). CCSA makes robots intrinsically m
Publikováno v:
Neural computation. 28(8)
Consider a self-motivated artificial agent who is exploring a complex environment. Part of the complexity is due to the raw high-dimensional sensory input streams, which the agent needs to make sense of. Such inputs can be compactly encoded through a
Publikováno v:
IJCNN
How can a humanoid robot autonomously learn and refine multiple sensorimotor skills as a byproduct of curiosity driven exploration, upon its high-dimensional unprocessed visual input? We present SKILLABILITY, which makes this possible. It combines th
Publikováno v:
From Animals to Animats 13 ISBN: 9783319088631
SAB
SAB
We present a motivational system for an agent undergoing reinforcement learning (RL), which enables it to balance multiple drives, each of which is satiated by different types of stimuli. Inspired by drive reduction theory, it uses Minor Component An
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a1576c7846e9e14bd0674fd679792fc7
https://doi.org/10.1007/978-3-319-08864-8_17
https://doi.org/10.1007/978-3-319-08864-8_17
Autor:
Varun Raj Kompella
In the absence of external guidance how can a robot learn to map the many raw pixels of high dimensional visual inputs to useful action sequences? I study methods that achieve this by making robots self motivated (curious) to continually build compac
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=snsf_p3_pubs::c44b1069538f16c72fffc62fa75a8afd
http://people.idsia.ch/~kompella/Papers/PhDThesis.pdf
http://people.idsia.ch/~kompella/Papers/PhDThesis.pdf
Publikováno v:
Frontiers in Neurorobotics, Vol 7 (2013)
Frontiers in Neurorobotics
Frontiers in Neurorobotics
Curiosity Driven Modular Incremental Slow Feature Analysis (CD-MISFA;~cite{cdmisfa}) is a recently introduced model of intrinsically-motivated invariance learning, which shows how curiosity enables the orderly formation of multiple stable sensory rep
Autonomous learning of abstractions using Curiosity-Driven Modular Incremental Slow Feature Analysis
Publikováno v:
ICDL-EPIROB
To autonomously learn behaviors in complex environments, vision-based agents need to develop useful sensory abstractions from high-dimensional video. We propose a modular, curiosity-driven learning system that autonomously learns multiple abstract re