HackerPapers

arXiv:1604.00289cs.AI

Building Machines That Learn and Think Like People

Brenden M. Lake, Tomer D. Ullman, Joshua B. Tenenbaum, Samuel J. Gershman

Abstract

Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many advances have come from using deep neural networks trained end-to-end in tasks such as object recognition, video games, and board games, achieving performance that equals or even beats humans in some respects. Despite their biological inspiration and performance achievements, these systems differ from human intelligence in crucial ways. We review progress in cognitive science suggesting that truly human-like learning and thinking machines will have to reach beyond current engineering trends in both what they learn, and how they learn it. Specifically, we argue that these machines should (a) build causal models of the world that support explanation and understanding, rather than merely solving pattern recognition problems; (b) ground learning in intuitive theories of physics and psychology, to support and enrich the knowledge that is learned; and (c) harness compositionality and learning-to-learn to rapidly acquire and generalize knowledge to new tasks and situations. We suggest concrete challenges and promising routes towards these goals that can combine the strengths of recent neural network advances with more structured cognitive models.

arXiv comments
In press at Behavioral and Brain Sciences. Open call for commentary proposals (until Nov. 22, 2016). https://www.cambridge.org/core/journals/behavioral-and-brain-sciences/information/calls-for-commentary/open-calls-for-commentary
Journal reference
DOI
Subjects
cs.AIArtificial Intelligence cs.CVComputer Vision and Pattern Recognition cs.LGMachine Learning cs.NENeural and Evolutionary Computing stat.MLMachine Learning (Statistics)

On Hacker News

Linked by 7 identifiable commenters in 9 comments across 7 threads, from Aug 2016 to Jun 2019. Found in 7 HN submissions from Apr 2016 to Apr 2020. Open submission discussions.

Comments linking this paper, per year.

Discussed in comments7 threads

  1. An Explicitly Relational Neural Network Architecture 1 comment

    • It depends how those symbols are encoded. Techniques like attention, and systems like transformers that are built on top of them, are often produce highly interpretable execution traces simply because their patterns of activity are very revealing of how they are going about…

      taliesinbView on HNParent context

  2. Automakers Are Rethinking the Timetable for Fully Autonomous Cars 1 comment

    • I have made this prediction multiple times throughout the last 8 years based on my knowledge of robotics. I have argued with family members who are VCs as well as enthusiastic technical observers based on simple first principles reasoning. It is highly unlikely that completely…

      areoformView on HNParent context

  3. The future of deep learning 1 comment

    • Counterfactuals in Pearl's sense are what allow particular models to be causal: to represent cause and effect under intervention, as opposed to mere correlation. This is an important part of how to build models that think like people[1]. [1] https://arxiv.org/abs/1604.00289

      eli_gottliebView on HNParent context

  4. How brains are built: Principles of computational neuroscience 1 comment

    • Hey guys, I found the non-crappy papers: https://arxiv.org/abs/1604.00289 -- Building Machines that Learn and Think like People http://rsif.royalsocietypublishing.org/content/13/122/201606... -- Active Inference and Robot Control: a case-study

      eli_gottliebView on HNParent context

  5. Ask HN: AI winter coming in next few years for AGI research? 1 comment

    • Its very hard to say as no one can really predict when a research drought will come unless you actually do the research. Though I must say that most of the ML researchers I've come across aren't directly concerned with AGI. (either they outright dont care, or they think its a…

      colorincorrectView on HNParent context

  6. Causality in machine learning 1 comment

    • I'd say that causality seems useful for intelligence. Josh Tenenbaum has pushed the idea of causal models and inference, and I find it compelling > Recent progress in artificial intelligence (AI) has renewed interest in building systems that learn and think like people. Many…

      pinouchonView on HNParent context

  7. The Difference Between AI, Machine Learning, and Deep Learning 3 comments

    • This difference is due to deep learning not being able to build models. Check out the recent paper "Building Machines That Learn and Think Like People": https://arxiv.org/abs/1604.00289 Humans can see one or a few examples of a novel object, such as a cat, and create a fully 3D…

      ansibleView on HNParent context

    • > Atari games just from the raw pixels on the screen It's important to distinguish between what sorts of games work well under this method and what sorts do not. Games that are variations of pole balancing, like Pong, fare better than more complex games like Asteroids, Frostbite…

      CybioteView on HNParent context

    • Some of these functions are easy to compose to get something impressive, true. But that is not often the case. Take the case for games and imagine we wanted a meta-algorithm to select an algorithm to apply to each game. The intelligence would then shift into, how does one select…

      CybioteView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.