Interleaving Fast and Slow Decision Making

The "Thinking, Fast and Slow" paradigm of Kahneman proposes that we use two different styles of thinking—a fast and intuitive System 1 for certain tasks, along with a slower but more analytical System 2 for others. While the idea of using this two-system style of thinking is gaining popularity in AI and robotics, our work considers how to interleave the two styles of decision-making, i.e., how System 1 and System 2 should be used together. For this, we propose a novel and general framework which includes a new System 0 to oversee Systems 1 and 2. At every point when a decision needs to be made, System 0 evaluates the situation and quickly hands over the decision-making process to either System 1 or System 2. We evaluate such a framework on a modified version of the classic Pac-Man game, with an already-trained RL algorithm for System 1, a Monte-Carlo tree search for System 2, and several different possible strategies for System 0. As expected, arbitrary switches between Systems 1 and 2 do not work, but certain strategies do well. With System 0, an agent is able to perform better than one that uses only System 1 or System 2.

[1]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[2]  Timothy W. Finin,et al.  Thinking, Fast and Slow: Combining Vector Spaces and Knowledge Graphs , 2017, ArXiv.

[3]  Giovanni Russo,et al.  Driving Reinforcement Learning with Models , 2020, IntelliSys.

[4]  Simon M. Lucas,et al.  Pac-Man Conquers Academia: Two Decades of Research Using a Classic Arcade Game , 2018, IEEE Transactions on Games.

[5]  Stuart J. Russell,et al.  Principles of Metareasoning , 1989, Artif. Intell..

[6]  Giovanni Pilato,et al.  Towards A Dual Process Approach to Computational Explanationon in Human-Robot Socia Interaction , 2017, CAID@IJCAI.

[7]  Marvin Minsky,et al.  Em-one: an architecture for reflective commonsense thinking , 2005 .

[8]  Honglak Lee,et al.  Deep Learning for Real-Time Atari Game Play Using Offline Monte-Carlo Tree Search Planning , 2014, NIPS.

[9]  Anita Raja,et al.  Metareasoning - Thinking about Thinking , 2011, Metareasoning.

[10]  Jason Teo,et al.  Application of artificial intelligence techniques in Ms. Pan-Man game : a review , 2015 .

[11]  Thomas L. Griffiths,et al.  When Does Bounded-Optimal Metareasoning Favor Few Cognitive Systems? , 2017, AAAI.

[12]  Adam Wierman,et al.  Thinking fast and slow: Optimization decomposition across timescales , 2017, 2017 IEEE 56th Annual Conference on Decision and Control (CDC).

[13]  J. Gero,et al.  EMPIRICAL EVIDENCE FOR KAHNEMAN ' S SYSTEM 1 AND SYSTEM 2 THINKING IN DESIGN , 2018 .

[14]  Ingmar Posner,et al.  Robots Thinking Fast and Slow: On Dual Process Theory and Metacognition in Embodied AI , 2020 .

[15]  Grady Booch,et al.  Thinking Fast and Slow in AI , 2020, ArXiv.

[16]  Ying Nian Wu,et al.  Multimodal Conditional Learning with Fast Thinking Policy-like Model and Slow Thinking Planner-like Model , 2019, ArXiv.

[17]  K. Stanovich,et al.  Heuristics and Biases: Individual Differences in Reasoning: Implications for the Rationality Debate? , 2002 .

[18]  David Barber,et al.  Thinking Fast and Slow with Deep Learning and Tree Search , 2017, NIPS.

[19]  Shu Wen Tay,et al.  Systems 1 and 2 thinking processes and cognitive reflection testing in medical students , 2016, Canadian medical education journal.

[20]  John DeNero,et al.  Teaching Introductory Artificial Intelligence with Pac-Man , 2010, Proceedings of the AAAI Conference on Artificial Intelligence.

[21]  Jaime F. Fisac,et al.  Planning, Fast and Slow: A Framework for Adaptive Real-Time Safe Trajectory Planning , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[22]  Marvin Minsky,et al.  Society of Mind: A Response to Four Reviews , 1991, Artif. Intell..

[23]  John M. Gregoire,et al.  Deep Reasoning Networks: Thinking Fast and Slow , 2019, ArXiv.

[24]  Giovanni Russo,et al.  Control-Tutored Reinforcement Learning , 2019, ArXiv.

[25]  G. Reeke.  Marvin Minsky, The Society of Mind , 1991, Artif. Intell..

[26]  D. Kahneman.  Thinking, Fast and Slow , 2011 .