Search Results for author:
Found 46 papers, 18 papers with code
Neural representation of action sequences: how far can a simple snippet-matching model take us?
The macaque Superior Temporal Sulcus (STS) is a brain area that receives and integrates inputs from both the ventral and dorsal visual processing streams (thought to specialize in form and motion processing respectively).
Neuronal Synchrony in Complex-Valued Deep Networks
Here, we show how aspects of spike timing, long hypothesized to play a crucial role in cortical information processing, could be incorporated into deep networks to build richer, versatile representations.
The Language of Actions: Recovering the Syntax and Semantics of Goal-Directed Human Activities
This paper describes a framework for modeling human activities as temporally structured processes.
Cooking in the kitchen: Recognizing and Segmenting Human Activities in Videos
Through extensive system evaluations, we demonstrate that combining compact video representations based on Fisher Vectors with HMM-based modeling yields very significant gains in accuracy and when properly trained with sufficient training samples, structured temporal models outperform unstructured bag-of-word types of models by a large margin on the tested performance metric.
An end-to-end generative framework for video segmentation and recognition
We describe an end-to-end generative approach for the segmentation and recognition of human activities.
How Deep is the Feature Analysis underlying Rapid Visual Categorization?
We found that recognition accuracy increases with higher stages of visual processing (higher level stages indeed outperforming human participants on the same task) but that human decisions agree best with predictions from intermediate stages.
What are the visual features underlying human versus machine vision?
Our study demonstrates that the narrowing gap between the object recognition accuracy of human observers and DCNs obscures distinct visual strategies used by each to achieve this performance.
Not-So-CLEVR: Visual Relations Strain Feedforward Neural Networks
The robust and efficient recognition of visual relations in images is a hallmark of biological vision.
Same-different problems strain convolutional neural networks
The robust and efficient recognition of visual relations in images is a hallmark of biological vision.
Learning long-range spatial dependencies with horizontal gated-recurrent units
As a prime example, convolutional neural networks, a type of feedforward neural networks, are now approaching -- and sometimes even surpassing -- human accuracy on a variety of visual recognition tasks.
Learning what and where to attend
Most recent gains in visual recognition have originated from the inclusion of attention mechanisms in deep convolutional networks (DCNs).
Robust pose tracking with a joint model of appearance and shape
We present a novel approach for estimating the 2D pose of an articulated object with an application to automated video analysis of small laboratory animals.
Robust neural circuit reconstruction from serial electron microscopy with convolutional recurrent networks
We first demonstrate that current state-of-the-art approaches to neuron segmentation perform poorly on the challenge.
Learning long-range spatial dependencies with horizontal gated recurrent units
Progress in deep learning has spawned great successes in many engineering applications.
Learning what and where to attend with humans in the loop
Most recent gains in visual recognition have originated from the inclusion of attention mechanisms in deep convolutional networks (DCNs).
Disentangling neural mechanisms for perceptual grouping
Forming perceptual groups and individuating objects in visual scenes is an essential step towards visual intelligence.
Stable and expressive recurrent vision models
We posit that the effectiveness of recurrent vision models is bottlenecked by the standard algorithm used for training them, "back-propagation through time" (BPTT), which has O(N) memory-complexity for training an N step model.
Go with the Flow: Adaptive Control for Neural ODEs
Despite their elegant formulation and lightweight memory cost, neural ordinary differential equations (NODEs) suffer from known representational limitations.
How Good is your Explanation? Algorithmic Stability Measures to Assess the Quality of Explanations for Deep Neural Networks
A plethora of methods have been proposed to explain how deep neural networks reach their decisions but comparatively, little effort has been made to ensure that the explanations produced by these methods are objectively relevant.
Recurrent neural circuits for contour detection
We introduce a deep recurrent neural network architecture that approximates visual cortical circuits.
Iterative VAE as a predictive brain model for out-of-distribution generalization
Our ability to generalize beyond training data to novel, out-of-distribution, image degradations is a hallmark of primate vision.
KuraNet: Systems of Coupled Oscillators that Learn to Synchronize
Here, we present a single approach to both of these problems in the form of "KuraNet", a deep-learning-based system of coupled oscillators that can learn to synchronize across a distribution of disordered network conditions.
Tracking Without Re-recognition in Humans and Machines
For this, we introduce PathTracker, a synthetic visual challenge that asks human observers and machines to track a target object in the midst of identical-looking "distractor" objects.
Understanding the computational demands underlying visual reasoning
Our analysis reveals a novel taxonomy of visual reasoning tasks, which can be primarily explained by both the type of relations (same-different vs. spatial-relation judgments) and the number of relations used to compose the underlying rules.
The Challenge of Appearance-Free Object Tracking with Feedforward Neural Networks
Here, we introduce $\textit{PathTracker}$, a visual challenge inspired by cognitive psychology, which tests the ability of observers to learn to track objects solely by their motion.
Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis
We describe a novel attribution method which is grounded in Sensitivity Analysis and uses Sobol indices.
What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methods
A multitude of explainability methods and associated fidelity performance metrics have been proposed to help better understand how modern AI systems make decisions.
Meta-Reinforcement Learning with Self-Modifying Networks
Deep Reinforcement Learning has demonstrated the potential of neural networks tuned with gradient descent for solving complex tasks in well-delimited environments.
Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysis
A variety of methods have been proposed to try to explain how deep neural networks make their decisions.
Diversity vs. Recognizability: Human-like generalization in one-shot generative models
Robust generalization to new concepts has long remained a distinctive feature of human intelligence.
Xplique: A Deep Learning Explainability Toolbox
Today's most advanced machine-learning models are hardly scrutable.
GAMR: A Guided Attention Model for (visual) Reasoning
Humans continue to outperform modern AI systems in their ability to flexibly parse and understand complex visual scenes.
A Benchmark for Compositional Visual Reasoning
Overall, we hope that our challenge will spur interest in the development of neural architectures that can learn to harness compositionality toward more efficient learning.
Conviformers: Convolutionally guided Vision Transformer
Vision transformers are nowadays the de-facto choice for image classification tasks.
Ranked #1 on
Fine-Grained Image Classification
on Herbarium 2022
Harmonizing the object recognition strategies of deep neural networks with humans
Across 84 different DNNs trained on ImageNet and three independent datasets measuring the where and the how of human visual strategies for object recognition on those images, we find a systematic trade-off between DNN categorization accuracy and alignment with human visual strategies for object recognition.
CRAFT: Concept Recursive Activation FacTorization for Explainability
However, recent research has exposed the limited practical value of these methods, attributed in part to their narrow focus on the most prominent regions of an image -- revealing "where" the model looks, but failing to elucidate "what" the model sees in those areas.
Break It Down: Evidence for Structural Compositionality in Neural Networks
Though modern neural networks have achieved impressive performance in both vision and language tasks, we know little about the functions that they implement.
Diffusion Models as Artists: Are we Closing the Gap between Humans and Machines?
An important milestone for AI is the development of algorithms that can produce drawings that are indistinguishable from those of humans.
Adversarial alignment: Breaking the trade-off between the strength of an attack and its relevance to human perception
Harmonized DNNs achieve the best of both worlds and experience attacks that are detectable and affect features that humans find diagnostic for recognition, meaning that attacks on these models are more likely to be rendered ineffective by inducing similar effects on human perception.
Unlocking Feature Visualization for Deeper Networks with MAgnitude Constrained Optimization
However, its widespread adoption has been limited due to a reliance on tricks to generate interpretable images, and corresponding challenges in scaling it to deeper neural networks.
Saliency strikes back: How filtering out high frequencies improves white-box explanations
Attribution methods correspond to a class of explainability methods (XAI) that aim to assess how individual inputs contribute to a model's decision-making process.
NeuroSurgeon: A Toolkit for Subnetwork Analysis
Despite recent advances in the field of explainability, much remains unknown about the algorithms that neural networks learn to represent.
Diagnosing and exploiting the computational demands of videos games for deep reinforcement learning
Humans learn by interacting with their environments and perceiving the outcomes of their actions.
Fixing the problems of deep neural networks will require better training data and learning algorithms
Bowers and colleagues argue that DNNs are poor models of biological vision because they often learn to rival human accuracy by relying on strategies that differ markedly from those of humans.
Uncovering Intermediate Variables in Transformers using Circuit Probing
We apply this method to models trained on simple arithmetic tasks, demonstrating its effectiveness at (1) deciphering the algorithms that models have learned, (2) revealing modular structure within a model, and (3) tracking the development of circuits over training.
Categorizing the Visual Environment and Analyzing the Visual Attention of Dogs
The MaskRCNN, with eye tracking apparatus, serves as an end to end model for automatically classifying the visual fixations of dogs.
