no code implementations • 20 Nov 2023 • Shreyas Sundara Raman, Madeline H. Pelgrim, Daphna Buchsbaum, Thomas Serre
The MaskRCNN, with eye tracking apparatus, serves as an end to end model for automatically classifying the visual fixations of dogs.
1 code implementation • 7 Nov 2023 • Michael A. Lepori, Thomas Serre, Ellie Pavlick
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.
no code implementations • 26 Sep 2023 • Drew Linsley, Thomas Serre
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.
no code implementations • 22 Sep 2023 • Lakshmi Narasimhan Govindarajan, Rex G Liu, Drew Linsley, Alekh Karkada Ashok, Max Reuter, Michael J Frank, Thomas Serre
Humans learn by interacting with their environments and perceiving the outcomes of their actions.
1 code implementation • 1 Sep 2023 • Michael A. Lepori, Ellie Pavlick, Thomas Serre
Despite recent advances in the field of explainability, much remains unknown about the algorithms that neural networks learn to represent.
no code implementations • 18 Jul 2023 • Sabine Muzellec, Thomas Fel, Victor Boutin, Léo Andéol, Rufin VanRullen, Thomas Serre
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.
1 code implementation • 11 Jun 2023 • Thomas Fel, Thibaut Boissin, Victor Boutin, Agustin Picard, Paul Novello, Julien Colin, Drew Linsley, Tom Rousseau, Rémi Cadène, Laurent Gardes, Thomas Serre
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.
no code implementations • 5 Jun 2023 • Drew Linsley, Pinyuan Feng, Thibaut Boissin, Alekh Karkada Ashok, Thomas Fel, Stephanie Olaiya, Thomas Serre
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.
1 code implementation • 27 Jan 2023 • Victor Boutin, Thomas Fel, Lakshya Singhal, Rishav Mukherji, Akash Nagaraj, Julien Colin, Thomas Serre
An important milestone for AI is the development of algorithms that can produce drawings that are indistinguishable from those of humans.
1 code implementation • NeurIPS 2023 • Michael A. Lepori, Thomas Serre, Ellie Pavlick
Though modern neural networks have achieved impressive performance in both vision and language tasks, we know little about the functions that they implement.
1 code implementation • CVPR 2023 • Thomas Fel, Agustin Picard, Louis Bethune, Thibaut Boissin, David Vigouroux, Julien Colin, Rémi Cadène, Thomas Serre
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.
3 code implementations • 8 Nov 2022 • Thomas Fel, Ivan Felipe, Drew Linsley, Thomas Serre
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.
1 code implementation • 17 Aug 2022 • Mohit Vaishnav, Thomas Fel, Ivań Felipe Rodríguez, Thomas Serre
Vision transformers are nowadays the de-facto choice for image classification tasks.
Ranked #1 on Fine-Grained Image Classification on Herbarium 2022
1 code implementation • 11 Jun 2022 • Aimen Zerroug, Mohit Vaishnav, Julien Colin, Sebastian Musslick, Thomas Serre
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.
1 code implementation • 10 Jun 2022 • Mohit Vaishnav, Thomas Serre
Humans continue to outperform modern AI systems in their ability to flexibly parse and understand complex visual scenes.
1 code implementation • 9 Jun 2022 • Thomas Fel, Lucas Hervier, David Vigouroux, Antonin Poche, Justin Plakoo, Remi Cadene, Mathieu Chalvidal, Julien Colin, Thibaut Boissin, Louis Bethune, Agustin Picard, Claire Nicodeme, Laurent Gardes, Gregory Flandin, Thomas Serre
Today's most advanced machine-learning models are hardly scrutable.
2 code implementations • 20 May 2022 • Victor Boutin, Lakshya Singhal, Xavier Thomas, Thomas Serre
Robust generalization to new concepts has long remained a distinctive feature of human intelligence.
no code implementations • CVPR 2023 • Thomas Fel, Melanie Ducoffe, David Vigouroux, Remi Cadene, Mikael Capelle, Claire Nicodeme, Thomas Serre
A variety of methods have been proposed to try to explain how deep neural networks make their decisions.
no code implementations • 4 Feb 2022 • Mathieu Chalvidal, Thomas Serre, Rufin VanRullen
Deep Reinforcement Learning has demonstrated the potential of neural networks tuned with gradient descent for solving complex tasks in well-delimited environments.
1 code implementation • 6 Dec 2021 • Julien Colin, Thomas Fel, Remi Cadene, Thomas Serre
A multitude of explainability methods and associated fidelity performance metrics have been proposed to help better understand how modern AI systems make decisions.
1 code implementation • NeurIPS 2021 • Thomas Fel, Remi Cadene, Mathieu Chalvidal, Matthieu Cord, David Vigouroux, Thomas Serre
We describe a novel attribution method which is grounded in Sensitivity Analysis and uses Sobol indices.
no code implementations • 30 Sep 2021 • Girik Malik, Drew Linsley, Thomas Serre, Ennio Mingolla
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.
no code implementations • 8 Aug 2021 • Mohit Vaishnav, Remi Cadene, Andrea Alamia, Drew Linsley, Rufin VanRullen, Thomas Serre
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.
no code implementations • NeurIPS 2021 • Drew Linsley, Girik Malik, Junkyung Kim, Lakshmi N Govindarajan, Ennio Mingolla, Thomas Serre
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.
1 code implementation • 6 May 2021 • Matthew Ricci, Minju Jung, Yuwei Zhang, Mathieu Chalvidal, Aneri Soni, Thomas Serre
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.
no code implementations • NeurIPS Workshop SVRHM 2020 • Victor Boutin, Aimen Zerroug, Minju Jung, Thomas Serre
Our ability to generalize beyond training data to novel, out-of-distribution, image degradations is a hallmark of primate vision.
no code implementations • ICLR 2020 • Drew Linsley, Junkyung Kim, Alekh Ashok, Thomas Serre
We introduce a deep recurrent neural network architecture that approximates visual cortical circuits.
no code implementations • 7 Sep 2020 • Thomas Fel, David Vigouroux, Rémi Cadène, Thomas Serre
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.
no code implementations • ICLR 2021 • Mathieu Chalvidal, Matthew Ricci, Rufin VanRullen, Thomas Serre
Despite their elegant formulation and lightweight memory cost, neural ordinary differential equations (NODEs) suffer from known representational limitations.
1 code implementation • NeurIPS 2020 • Drew Linsley, Alekh Karkada Ashok, Lakshmi Narasimhan Govindarajan, Rex Liu, Thomas Serre
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.
no code implementations • ICLR 2020 • Junkyung Kim, Drew Linsley, Kalpit Thakkar, Thomas Serre
Forming perceptual groups and individuating objects in visual scenes is an essential step towards visual intelligence.
no code implementations • ICLR 2019 • Drew Linsley, Dan Shiebler, Sven Eberhardt, Thomas Serre
Most recent gains in visual recognition have originated from the inclusion of attention mechanisms in deep convolutional networks (DCNs).
no code implementations • NeurIPS 2018 • Drew Linsley, Junkyung Kim, Vijay Veerabadran, Charles Windolf, Thomas Serre
Progress in deep learning has spawned great successes in many engineering applications.
no code implementations • 28 Nov 2018 • Drew Linsley, Junkyung Kim, David Berson, Thomas Serre
We first demonstrate that current state-of-the-art approaches to neuron segmentation perform poorly on the challenge.
no code implementations • 28 Jun 2018 • Yuliang Guo, Lakshmi Narasimhan Govindarajan, Benjamin Kimia, Thomas Serre
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.
1 code implementation • 22 May 2018 • Drew Linsley, Dan Shiebler, Sven Eberhardt, Thomas Serre
Most recent gains in visual recognition have originated from the inclusion of attention mechanisms in deep convolutional networks (DCNs).
1 code implementation • NeurIPS 2018 • Drew Linsley, Junkyung Kim, Vijay Veerabadran, Thomas Serre
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.
no code implementations • 9 Feb 2018 • Matthew Ricci, Junkyung Kim, Thomas Serre
The robust and efficient recognition of visual relations in images is a hallmark of biological vision.
no code implementations • ICLR 2018 • Junkyung Kim, Matthew Ricci, Thomas Serre
The robust and efficient recognition of visual relations in images is a hallmark of biological vision.
no code implementations • 10 Jan 2017 • Drew Linsley, Sven Eberhardt, Tarun Sharma, Pankaj Gupta, Thomas Serre
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.
no code implementations • NeurIPS 2016 • Sven Eberhardt, Jonah Cader, Thomas Serre
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.
no code implementations • 7 Sep 2015 • Hilde Kuehne, Juergen Gall, Thomas Serre
We describe an end-to-end generative approach for the segmentation and recognition of human activities.
no code implementations • 25 Aug 2015 • Hilde Kuehne, Juergen Gall, Thomas Serre
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.
no code implementations • CVPR 2014 • Hilde Kuehne, Ali Arslan, Thomas Serre
This paper describes a framework for modeling human activities as temporally structured processes.
no code implementations • 20 Dec 2013 • David P. Reichert, Thomas Serre
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.
no code implementations • NeurIPS 2013 • Cheston Tan, Jedediah M. Singer, Thomas Serre, David Sheinberg, Tomaso Poggio
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).