Fast Slow Planning

About Fast and Slow AI: The newly introduced idea of fast and slow AI envisages a multi-agent AI architecture where incoming problems are solved by either system 1 (or ”fast” - S1) agents, also called solvers, that react by exploiting only past experience, or by system 2 (or ”slow” - S2) agents, that are deliberately activated when there is the need to reason and search for optimal solutions beyond what is expected from the system 1 agent (AAAI 2021).

Unique to Fast and Slow Planning: We are specifically investigating the nature of classical and epistemic planning in this context using a common metacognitive engine and planning-specific S1 solving and integration strategies.

Partners: IBM Research, Tulane University, Union College, University of Brescia, University of Udine, University of South Carolina, University of West Florida

Papers

• • Marianna Bergamaschi Ganapini, Murray Campbell, Francesco Fabiano, Lior Horesh, Jon Lenchner, Andrea Loreggia, Nicholas Mattei, Taher Rahgooy, Francesca Rossi, Biplav Srivastava, Kristen Brent Venable, Combining Fast and Slow Thinking for Human-like and Efficient Navigation in Constrained Environments. Preprint on Arxiv at: https://arxiv.org/abs/2201.07050, 2022 [Neuro-Symbolic AI, Metacognition]

  • Grady Booch, Francesco Fabiano, Lior Horesh, Kiran Kate, Jon Lenchner, Nick Linck, Andrea Loreggia, Keerthiram Murugesan, Nicholas Mattei, Francesca Rossi, Biplav Srivastava, Thinking Fast and Slow in AI, AAAI 2021. Preprint on Arxiv at: https://arxiv.org/abs/2010.06002 [Neuro-Symbolic AI]

  • Marianna Bergamaschi Ganapini, Murray Campbell, Francesco Fabiano, Lior Horesh, Jon Lenchner, Andrea Loreggia, Nicholas Mattei, Francesca Rossi, Biplav Srivastava, Kristen Brent Venable, Thinking Fast and Slow in AI: the Role of Metacognition. Preprint on Arxiv at: https://arxiv.org/abs/2110.01834, 2021 [Neuro-Symbolic AI, Metacognition]

  • Francesco Fabiano, Biplav Srivastava, Jonathan Lenchner, Lior Horesh, Francesca Rossi, Marianna Bergamaschi Ganapini, E-PDDL: A Standardized Way of Defining Epistemic Planning Problems. Workshop on Knowledge Engineering for Planning and Scheduling (KEPS), International Conference on Automated Planning and Scheduling (ICAPS), 2021. Preprint on Arxiv at: https://arxiv.org/abs/2107.08739. [Planning, Knowledge Representation]