Combining Fast and Slow Thinking for Human-like and Efficient Navigation in Constrained Environments

Current AI systems lack several important human capabilities, such as adaptability, generalizability, self-control, consistency, common sense, and causal reasoning. We believe that existing cognitive theories of human decision making, such as the thinking fast and slow theory, can provide insights on how to advance AI systems towards some of these capabilities. In this paper, we propose a general architecture that is based on fast/slow solvers and a metacognitive component. We then present experimental results on the behavior of an instance of this architecture, for AI systems that make decisions about navigating in a constrained environment. We show how combining the fast and slow decision modalities allows the system to evolve over time and gradually pass from slow to fast thinking with enough experience, and that this greatly helps in decision quality, resource consumption, and efficiency.

[1]  P. Carruthers.  Explicit nonconceptual metacognition , 2020, Philosophical Studies.

[2]  Yoshua Bengio,et al.  The Consciousness Prior , 2017, ArXiv.

[3]  Jonathan D. Cohen,et al.  The Expected Value of Control: An Integrative Theory of Anterior Cingulate Cortex Function , 2013, Neuron.

[4]  J. Flavell.  Metacognition and Cognitive Monitoring: A New Area of Cognitive-Developmental Inquiry. , 1979 .

[5]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

[6]  Francesca Rossi,et al.  Building Ethically Bounded AI , 2018, AAAI.

[7]  Anna Papst,et al.  Fast and Slow , 2008, Science.

[8]  Remo Guidieri.  Res , 1995, RES: Anthropology and Aesthetics.

[9]  T. O. Nelson.  Metamemory: A Theoretical Framework and New Findings , 1990 .

[10]  Rakefet Ackerman,et al.  Meta-Reasoning: Monitoring and Control of Thinking and Reasoning , 2017, Trends in Cognitive Sciences.

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

[12]  Sang Wan Lee,et al.  Task complexity interacts with state-space uncertainty in the arbitration between model-based and model-free learning , 2019, Nature Communications.

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

[14]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 1998, IEEE Trans. Neural Networks.

[15]  Michael I. Jordan,et al.  Advances in Neural Information Processing Systems 30 , 1995 .

[16]  J. Proust,et al.  The Philosophy of Metacognition: Mental Agency and Self-Awareness , 2013 .

[17]  Kush R. Varshney,et al.  Teaching AI Agents Ethical Values Using Reinforcement Learning and Policy Orchestration , 2019, IJCAI.

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

[19]  Jiyun Li,et al.  Theoretical developments in decision field theory: comment on Tsetsos, Usher, and Chater (2010). , 2010, Psychological review.

[20]  Gary Marcus,et al.  The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence , 2020, ArXiv.

[21]  Heike Trautmann,et al.  Automated Algorithm Selection: Survey and Perspectives , 2018, Evolutionary Computation.

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

[23]  Lee Sael,et al.  Procedia Computer Science , 2015 .

[24]  Shrisha Rao,et al.  Interleaving Fast and Slow Decision Making , 2020, 2021 IEEE International Conference on Robotics and Automation (ICRA).

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

[26]  J. Townsend,et al.  Decision field theory: a dynamic-cognitive approach to decision making in an uncertain environment. , 1993, Psychological review.

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

[28]  Valerie A. Thompson,et al.  Intuition, reason, and metacognition , 2011, Cognitive Psychology.

[29]  Michael T. Cox.  Metacognition in computation: A selected research review , 2005, Artif. Intell..

[30]  R. Ratcliff,et al.  Multialternative decision field theory: a dynamic connectionist model of decision making. , 2001, Psychological review.

[31]  Gustav Bergmann,et al.  PHILOSOPHICAL STUDIES , 1927, Philosophy.