Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning
By Kelsey R. Allen, Kevin A. Smith, Josh Tenenbaum. 2019
152 citations recorded by the source corpus.
Many animals, and an increasing number of artificial agents, display sophisticated capabilities to perceive and manipulate objects. But human beings remain distinctive in their capacity for flexible, creative tool use—using objects in new ways to act on the world, achieve a goal, or solve a problem. To study this type of general physical problem solving, we introduce the Virtual Tools game. In this game, people solve a large range of challenging physical puzzles in just a handful of attempts. We propose that the flexibility of human physical problem solving rests on an ability to imagine the effects of hypothesized actions, while the efficiency of human search arises from rich action priors which are updated via observations of the world. We instantiate these components in the “sample, simulate, update” (SSUP) model and show that it captures human performance across 30 levels of the Virtual Tools game. More broadly, this model provides a mechanism for explaining how people condense general physical knowledge into actionable, task-specific plans to achieve flexible and efficient physical problem solving.
Subfields: Artificial Intelligence, Control and Systems Engineering
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Joelle Pineau
- Josh Tenenbaum
- Sergey Levine
- Li Fei-Fei
- Yoshua Bengio
- Mohit Bansal
- Amy Zhang
- Ruslan Salakhutdinov
In Control and Systems Engineering
- Joelle Pineau
- Josh Tenenbaum
- Amy Zhang
- Daniel L. K. Yamins
- Damian Mrowca
- Shagun Sodhani
- Kevin A. Smith
- Li Fei-Fei
Citations in the corpus
- Tool Learning with Foundation Models (2023, unknown): “Alternatively, humans learn from trial and error to correct and rectify their tool-use behaviors (Allen et al., 2019).”
- Multi-Task Reinforcement Learning with Context-based Representations (2021, background): “In contrast, humans are adept at multi-tasking by acquiring new skills and composing known skills to solve complex tasks (Parascan-dolo et al., 2018; Allen et al., 2020).”
- Physion: Evaluating Physical Prediction from Vision in Humans and Machines (2021, background): “Datasets with greater heterogeneity in scenarios are designed in 2D with simple visual representations and are meant to test simple action planning abilities that rely on physics, rather than understanding of physical principles themselves [32, 6, 3].”