Sources of uncertainty in intuitive physics

By Kevin A. Smith, E. Vul. 2012

147 citations recorded by the source corpus.

Subfields: Developmental and Educational Psychology, Experimental and Cognitive Psychology, General Decision Sciences

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Developmental and Educational Psychology

In Experimental and Cognitive Psychology

In General Decision Sciences

Citations in the corpus

Flexible Neural Representation for Physics Prediction (2018, background): “Although human intuitive physics may be quantitatively inaccurate under some circumstances [32], humans make qualitatively plausible guesses about dynamic trajectories of their environments over long time horizons [41].”
A Compositional Object-Based Approach to Learning Physical Dynamics (2016, background): “While the NP and LSTM fail to predict plausible physical movement entirely, the NPE’s predictions initially adhere closely to the ground truth, then slowly diverge due to the accumulation of subtle errors, just as the human perceptual system also accumulates errors (Smith and Vul, 2013).”
Consistent physics underlying ballistic motion prediction (2013, background): “…predictions using quantitative cognitive models that assume people have an accurate and consistent understanding of the laws of physics that they apply flexibly across tasks (Hamrick, Battaglia, & Tenenbaum, 2011; Sanborn, Mansinghka, & Griffiths, 2013; Smith & Vul, 2013; Téglás et al., 2011).”
Reconciling intuitive physics and Newtonian mechanics for colliding objects. (2013, methodology): “This approach, which here we call noisy Newton, was introduced in Sanborn, Mansinghka, and Griffiths (2009) and has been subsequently applied to other aspects of intuitive physics (Gerstenberg, Goodman, Lagnado, & Tenenbaum, 2012; Hamrick, Battaglia, & Tenenbaum, 2011; Smith & Vul, 2012).”
PIP: Physical Interaction Prediction via Mental Imagery with Span Selection (2021, methodology): “We use noisy and approximate physical simulations to account for property, perceptual, dynamic and even collision uncertainties [15, 22, 23, 24, 25, 26, 17].”
Intuitive Theories (2020, background): “Smith and Vul (2013) set out to investigate more closely what sorts of noise in people's mental physical simulations best explains their actions in a simple physics game.”
A Comparative Evaluation of Approximate Probabilistic Simulation and Deep Neural Networks as Accounts of Human Physical Scene Understanding (2016, background): “These simulation engines approximate object dynamics interacting under Newtonian or other forms of classical mechanics over short time scales, in ways that are perceptually reasonable (if not necessarily physically accurate) and efﬁcient enough to run in real time for complex scenes.”
SCAN: Learning Hierarchical Compositional Visual Concepts (2017, background): “Intelligent agents are able to discover and learn abstract compositional concepts using little supervision (Baillargeon, 1987; Spelke, 1990; Baillargeon, 2004; Smith & Vul, 2013).”
How, whether, why: Causal judgments as counterfactual contrasts (2015, background): “In cases in which the ball made no difference to the actual outcome (i.e. PDM = 0), the other predictors were capped at 0.”
Small Sample Learning in Big Data Era (2018, background): “, 2015), common sense (Davis and Marcus, 2015), and intuitive physics (Smith and Vul, 2013; Battaglia et al., 2016; Hamrick et al., 2017)), and so on.”
Concepts in a Probabilistic Language of Thought (2014, background): “Recent research has shown that people’s intuitive physical judgments in several domains are well described by such noisy Newtonian simulations (Battaglia, Hamrick, & Tenenbaum, 2013; Sanborn, Mansinghka, & Griffiths, 2013; K. Smith, Dechter, Tenenbaum, & Vul, 2013; K. A. Smith & Vul, 2012).”
The Tools Challenge: Rapid Trial-and-Error Learning in Physical Problem Solving (2019, background): “This engine is able to simulate the world forwards in time with approximately correct but stochastic dynamics (Smith & Vul, 2013; Smith, Battaglia, & Vul, 2018).”
Rapid trial-and-error learning with simulation supports flexible tool use and physical reasoning (2019, background): “People characteristically have noisy predictions of how collisions will resolve (16), and so for simplicity we assume uncertainty about outcomes is driven only by noise in those collisions (the direction and amount of force that is applied between two colliding objects).”
Mind Games: Game Engines as an Architecture for Intuitive Physics. (2017, background): “…engines support a variety of predictions across di ﬀ erent tasks and types of reasoning, including predicting how a scene will unfold over time [6, 74], interacting with a dynamic scene [75], reasoning about underlying physical properties [10, 11, 66], causal judgments [7], and quantitative…”
From counterfactual simulation to causal judgment (2014, background): “By adding Gaussian noise to B's velocity vector , we capture the dynamic uncertainty inherent in people's intuitive physical model (cf. Smith & Vul, 2012).”