• Publications
  • Influence
The Large-Scale Structure of Semantic Networks
  • 65
  • 11
  • PDF
ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models
TLDR
We collect a large real-world test set, ObjectNet, for object recognition with controls where object backgrounds, rotations, and imaging viewpoints are random. Expand
  • 61
  • 10
  • PDF
The mentalistic basis of core social cognition: experiments in preverbal infants and a computational model.
Evaluating individuals based on their pro- and anti-social behaviors is fundamental to successful human interaction. Recent research suggests that even preverbal infants engage in social evaluation;Expand
  • 141
  • 2
  • PDF
Surprise! Infants consider possible bases of generalization for a single input example.
TLDR
We familiarized 9-month-olds with a single three-syllable input example that contained either one surprising feature (repetition and a rare syllable, Experiment 1) or two features (re repetitions and rare syllables, Experiment 2). Expand
  • 31
  • 2
  • PDF
Modeling Expectation Violation in Intuitive Physics with Coarse Probabilistic Object Representations
TLDR
We propose ADEPT, a model that uses a coarse (approximate geometry) object-centric representation for dynamic 3D scene understanding. Expand
  • 12
  • 2
  • PDF
McGovern Institute for Brain Research
The McGovern Institute for Brain Research at MIT is led by a team of world-renowned neuroscientists committed to meeting two great challenges of modern science: understanding how the brain works andExpand
  • 2
  • 1
  • PDF
A stochastic programming perspective on nonparametric Bayes
We use Church, a Turing-universal language for stochastic generative processes and the probability distributions they induce, to study and extend several objects in nonparametric Bayesian statistics.Expand
  • 17
  • PDF
Efficient inverse graphics in biological face processing
TLDR
Neural networks in the primate brain may invert a graphics style model of how 3D object shapes and textures cause observed images. Expand
  • 4
Visual Concept-Metaconcept Learning
TLDR
We propose the visual concept-metaconcept learner (VCML) for joint learning of concepts and metaconcepts from images and associated question-answer pairs. Expand
  • 8
  • PDF