agents   2609

« earlier    

[1802.07740] Machine Theory of Mind
Theory of mind (ToM; Premack & Woodruff, 1978) broadly refers to humans' ability to represent the mental states of others, including their desires, beliefs, and intentions. We propose to train a machine to build such models too. We design a Theory of Mind neural network -- a ToMnet -- which uses meta-learning to build models of the agents it encounters, from observations of their behaviour alone. Through this process, it acquires a strong prior model for agents' behaviour, as well as the ability to bootstrap to richer predictions about agents' characteristics and mental states using only a small number of behavioural observations. We apply the ToMnet to agents behaving in simple gridworld environments, showing that it learns to model random, algorithmic, and deep reinforcement learning agents from varied populations, and that it passes classic ToM tasks such as the "Sally-Anne" test (Wimmer & Perner, 1983; Baron-Cohen et al., 1985) of recognising that others can hold false beliefs about the world. We argue that this system -- which autonomously learns how to model other agents in its world -- is an important step forward for developing multi-agent AI systems, for building intermediating technology for machine-human interaction, and for advancing the progress on interpretable AI.
artificial-intelligence  relevance  machine-learning  to-write-about  collective-intelligence  agents  consider:the-mangle 
23 days ago by Vaguery

« earlier    

related tags

/  (sight  (so  -  -  2017-07-25  2017-07-27  2017-07-29  2018  3d  594  628  66.249.7…  741  741…  869  a...  a.  academic  agency  agencys  agent  aggregators  agriculture  ai  and  andre  angeles  animation  api  architecture  art  articles  artificial-intelligence  attracting  automation  bad  ball  bayesian  behavior  blacklisted  block  blockchain  blogger:  body  boids  book  books  bots  brokers  business  canad  chatbot  cognition  collective-intelligence  collective  computer+science  connecting  consider:the-mangle  constraints  consultants  consum  conversational  conveyancing  corporate  countries  crawlers  creativity  crowd  daemons  deep  deepmind  development  dialogflow  disable  docs  drones  economics  editing  english  espn  estate  facebook  facilitators  far)  far  farming  finance  financial  for  form  forms  frameworks  france  free  gamedesign  gameresearch  games  generators  germany  github  golden  good  google  heterogeneous  hipchat  hosts  humana  ifttt  iguodala  imagine  inequality  infovis  investment  iot  ip's  ip  ips  is  javascript  jenkins  kent  keynesian  kindle  la  lakers  language  languages  leadership  liquidity  livingst...  lonzo  los  luxury  machine-learning  machine.learning  machinelearning  mackenzie  macos  maldata  malicious  mathematics  matterport  meet  miamibeach  mind  mix  monetary  monitoring  multiagent  navigation  navmesh  networks  neuralnets  new  nlp  nonfiction  nouvelle-aquitaine  object  on  opensource  or  our  partner  partners  pathfinding  pau  photography  pinterest  plan  policy  pomdp  probabilistic-programming  probabilistic_programming  probability  product  products  programming  property  ption  publisher  publishing  pyrénées-atlantiques  python  querying  rails  ramsgate  real-estate  realestate  realtors  recsys  reinforcement+learning  reinforcement-learning  reinforcement_learning  relevance  renting  retail  reviews  robot  rss  scout  search  services  shaun  shiny  should  simulation  simulations  sketching17  slight  smarttechnology  smith...  software  spain  spiders  sport  sportscenter  sportsnation  state  statistics  steering  stephen  storytelling  straight  substitution_of_humans  sysadmin  team?  techno  tensorflow  text  the  titaa  to-write-about  tools  top  tops  triangulation  trolling?  ukraine  unable  update)  updates)  user  voice  vr  warriora  warriors  web  who  woodland  wordpress  writing  writing_business  you  |   

Copy this bookmark: