Matches in SemOpenAlex for { <https://semopenalex.org/work/W3113184484> ?p ?o ?g. }
- W3113184484 abstract "A common vision from science fiction is that robots will one day inhabit our physical spaces, sense the world as we do, assist our physical labours, and communicate with us through natural language. Here we study how to design artificial agents that can interact naturally with humans using the simplification of a virtual environment. This setting nevertheless integrates a number of the central challenges of artificial intelligence (AI) research: complex visual perception and goal-directed physical control, grounded language comprehension and production, and multi-agent social interaction. To build agents that can robustly interact with humans, we would ideally train them while they interact with humans. However, this is presently impractical. Therefore, we approximate the role of the human with another learned agent, and use ideas from inverse reinforcement learning to reduce the disparities between human-human and agent-agent interactive behaviour. Rigorously evaluating our agents poses a great challenge, so we develop a variety of behavioural tests, including evaluation by humans who watch videos of agents or interact directly with them. These evaluations convincingly demonstrate that interactive training and auxiliary losses improve agent behaviour beyond what is achieved by supervised learning of actions alone. Further, we demonstrate that agent capabilities generalise beyond literal experiences in the dataset. Finally, we train evaluation models whose ratings of agents agree well with human judgement, thus permitting the evaluation of new agent models without additional effort. Taken together, our results in this virtual environment provide evidence that large-scale human behavioural imitation is a promising tool to create intelligent, interactive agents, and the challenge of reliably evaluating such agents is possible to surmount." @default.
- W3113184484 created "2020-12-21" @default.
- W3113184484 creator A5005420880 @default.
- W3113184484 creator A5010396788 @default.
- W3113184484 creator A5013028446 @default.
- W3113184484 creator A5014071624 @default.
- W3113184484 creator A5024479338 @default.
- W3113184484 creator A5026344718 @default.
- W3113184484 creator A5033390440 @default.
- W3113184484 creator A5035099870 @default.
- W3113184484 creator A5039232395 @default.
- W3113184484 creator A5039426831 @default.
- W3113184484 creator A5040662871 @default.
- W3113184484 creator A5044159906 @default.
- W3113184484 creator A5044385855 @default.
- W3113184484 creator A5044961078 @default.
- W3113184484 creator A5046028804 @default.
- W3113184484 creator A5052084048 @default.
- W3113184484 creator A5054522787 @default.
- W3113184484 creator A5056029139 @default.
- W3113184484 creator A5059282957 @default.
- W3113184484 creator A5059348157 @default.
- W3113184484 creator A5066294254 @default.
- W3113184484 creator A5068452444 @default.
- W3113184484 creator A5069438696 @default.
- W3113184484 creator A5069796663 @default.
- W3113184484 creator A5070090646 @default.
- W3113184484 creator A5080601982 @default.
- W3113184484 creator A5081473628 @default.
- W3113184484 creator A5083387921 @default.
- W3113184484 creator A5089917436 @default.
- W3113184484 date "2020-12-10" @default.
- W3113184484 modified "2023-09-27" @default.
- W3113184484 title "Imitating Interactive Intelligence" @default.
- W3113184484 cites W107583932 @default.
- W3113184484 cites W1482083705 @default.
- W3113184484 cites W1506955096 @default.
- W3113184484 cites W1536680647 @default.
- W3113184484 cites W1810943226 @default.
- W3113184484 cites W1991691398 @default.
- W3113184484 cites W2005814556 @default.
- W3113184484 cites W2010916807 @default.
- W3113184484 cites W2037530582 @default.
- W3113184484 cites W2079145130 @default.
- W3113184484 cites W2099471712 @default.
- W3113184484 cites W2103702476 @default.
- W3113184484 cites W2137391072 @default.
- W3113184484 cites W2145482038 @default.
- W3113184484 cites W2150671164 @default.
- W3113184484 cites W2152790380 @default.
- W3113184484 cites W2157364932 @default.
- W3113184484 cites W2158604749 @default.
- W3113184484 cites W2167224731 @default.
- W3113184484 cites W2194775991 @default.
- W3113184484 cites W2236233024 @default.
- W3113184484 cites W2257979135 @default.
- W3113184484 cites W2260190802 @default.
- W3113184484 cites W2474463024 @default.
- W3113184484 cites W2566467060 @default.
- W3113184484 cites W2591957724 @default.
- W3113184484 cites W2606722458 @default.
- W3113184484 cites W2627585944 @default.
- W3113184484 cites W2698595662 @default.
- W3113184484 cites W2729615412 @default.
- W3113184484 cites W2774005037 @default.
- W3113184484 cites W2786036274 @default.
- W3113184484 cites W2794908222 @default.
- W3113184484 cites W2842511635 @default.
- W3113184484 cites W2944828972 @default.
- W3113184484 cites W2952581030 @default.
- W3113184484 cites W2962715211 @default.
- W3113184484 cites W2962957031 @default.
- W3113184484 cites W2963150697 @default.
- W3113184484 cites W2963277051 @default.
- W3113184484 cites W2963328631 @default.
- W3113184484 cites W2963341956 @default.
- W3113184484 cites W2963403868 @default.
- W3113184484 cites W2963800628 @default.
- W3113184484 cites W2963921132 @default.
- W3113184484 cites W2963925437 @default.
- W3113184484 cites W2964043796 @default.
- W3113184484 cites W2964121744 @default.
- W3113184484 cites W2964263543 @default.
- W3113184484 cites W2970469901 @default.
- W3113184484 cites W2977505966 @default.
- W3113184484 cites W2982316857 @default.
- W3113184484 cites W2989818084 @default.
- W3113184484 cites W2990152177 @default.
- W3113184484 cites W2994803089 @default.
- W3113184484 cites W3001279689 @default.
- W3113184484 cites W3004125788 @default.
- W3113184484 cites W3004815632 @default.
- W3113184484 cites W3005188920 @default.
- W3113184484 cites W3012990076 @default.
- W3113184484 cites W3025552214 @default.
- W3113184484 cites W3028821797 @default.
- W3113184484 cites W3030163527 @default.
- W3113184484 cites W3047499204 @default.
- W3113184484 cites W3084747096 @default.
- W3113184484 cites W3102497351 @default.