Search Results for author:
Found 22 papers, 13 papers with code
Unsupervised Segmentation in Real-World Images via Spelke Object Inference
Self-supervised, category-agnostic segmentation of real-world images is a challenging open problem in computer vision.
Limiting Dynamics of SGD: Modified Loss, Phase Space Oscillations, and Anomalous Diffusion
In this work we explore the limiting dynamics of deep neural networks trained with stochastic gradient descent (SGD).
Physion: Evaluating Physical Prediction from Vision in Humans and Machines
While current vision algorithms excel at many challenging tasks, it is unclear how well they understand the physical dynamics of real-world environments.
The ThreeDWorld Transport Challenge: A Visually Guided Task-and-Motion Planning Benchmark for Physically Realistic Embodied AI
To complete the task, an embodied agent must plan a sequence of actions to change the state of a large number of objects in the face of realistic physical constraints.
Neural Mechanics: Symmetry and Broken Conservation Laws in Deep Learning Dynamics
Overall, by exploiting symmetry, our work demonstrates that we can analytically describe the learning dynamics of various parameter combinations at finite learning rates and batch sizes for state of the art architectures trained on any dataset.
Identifying Learning Rules From Neural Network Observables
We show that different classes of learning rules can be separated solely on the basis of aggregate statistics of the weights, activations, or instantaneous layer-wise activity changes, and that these results generalize to limited access to the trajectory and held-out architectures and learning curricula.
ThreeDWorld: A Platform for Interactive Multi-Modal Physical Simulation
We introduce ThreeDWorld (TDW), a platform for interactive multi-modal physical simulation.
Learning Physical Graph Representations from Visual Scenes
To overcome these limitations, we introduce the idea of Physical Scene Graphs (PSGs), which represent scenes as hierarchical graphs, with nodes in the hierarchy corresponding intuitively to object parts at different scales, and edges to physical connections between parts.
Pruning neural networks without any data by iteratively conserving synaptic flow
Pruning the parameters of deep neural networks has generated intense interest due to potential savings in time, memory and energy both during training and at test time.
Visual Grounding of Learned Physical Models
The abilities to perform physical reasoning and to adapt to new environments, while intrinsic to humans, remain challenging to state-of-the-art computational models.
Two Routes to Scalable Credit Assignment without Weight Symmetry
The neural plausibility of backpropagation has long been disputed, primarily for its use of non-local weight transport $-$ the biologically dubious requirement that one neuron instantaneously measure the synaptic weights of another.
Brain-Score: Which Artificial Neural Network for Object Recognition is most Brain-Like?
We therefore developed Brain-Score – a composite of multiple neural and behavioral benchmarks that score any ANN on how similar it is to the brain’s mechanisms for core object recognition – and we deployed it to evaluate a wide range of state-of-the-art deep ANNs.
Stochastic Neural Physics Predictor
Recently, neural-network based forward dynamics models have been proposed that attempt to learn the dynamics of physical systems in a deterministic way.
Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs
Deep convolutional artificial neural networks (ANNs) are the leading class of candidate models of the mechanisms of visual processing in the primate ventral stream.
Aligning Artificial Neural Networks to the Brain yields Shallow Recurrent Architectures
Deep artificial neural networks with spatially repeated processing (a. k. a., deep convolutional ANNs) have been established as the best class of candidate models of visual processing in the primate ventral visual processing stream.
Flexible Neural Representation for Physics Prediction
Humans have a remarkable capacity to understand the physical dynamics of objects in their environment, flexibly capturing complex structures and interactions at multiple levels of detail.
Task-Driven Convolutional Recurrent Models of the Visual System
Feed-forward convolutional neural networks (CNNs) are currently state-of-the-art for object classification tasks such as ImageNet.
Learning to Play with Intrinsically-Motivated Self-Aware Agents
We demonstrate that this policy causes the agent to explore novel and informative interactions with its environment, leading to the generation of a spectrum of complex behaviors, including ego-motion prediction, object attention, and object gathering.
Emergence of Structured Behaviors from Curiosity-Based Intrinsic Motivation
Moreover, the world model that the agent learns supports improved performance on object dynamics prediction and localization tasks.
Modular Continual Learning in a Unified Visual Environment
A core aspect of human intelligence is the ability to learn new tasks quickly and switch between them flexibly.
A Useful Motif for Flexible Task Learning in an Embodied Two-Dimensional Visual Environment
Recent results from neuroscience and artificial intelligence suggest the role of the general purpose visual representation may be played by a deep convolutional neural network, and give some clues how task modules based on such a representation might be discovered and constructed.
Deep Neural Networks Rival the Representation of Primate IT Cortex for Core Visual Object Recognition
Our evaluations show that, unlike previous bio-inspired models, the latest DNNs rival the representational performance of IT cortex on this visual object recognition task.
