Drew Linsley
YOU?
Author Swipe
View article: Better artificial intelligence does not mean better models of biology
Better artificial intelligence does not mean better models of biology Open
Deep neural networks (DNNs) once showed increasing alignment with primate perception and neural responses as they improved on vision benchmarks, raising hopes that advances in AI would yield better models of biological vision. However, we …
View article: Tracking objects that change in appearance with phase synchrony
Tracking objects that change in appearance with phase synchrony Open
Objects we encounter often change appearance as we interact with them. Changes in illumination (shadows), object pose, or the movement of non-rigid objects can drastically alter available image features. How do biological visual systems tr…
View article: Building better models of biological vision by searching for more ecological data diets and learning objectives
Building better models of biological vision by searching for more ecological data diets and learning objectives Open
International audience
View article: The 3D-PC: a benchmark for visual perspective taking in humans and machines
The 3D-PC: a benchmark for visual perspective taking in humans and machines Open
Visual perspective taking (VPT) is the ability to perceive and reason about the perspectives of others. It is an essential feature of human intelligence, which develops over the first decade of life and requires an ability to process the 3…
View article: Deceptive learning in histopathology
Deceptive learning in histopathology Open
Aims Deep learning holds immense potential for histopathology, automating tasks that are simple for expert pathologists and revealing novel biology for tasks that were previously considered difficult or impossible to solve by eye alone. Ho…
View article: Neural scaling laws for phenotypic drug discovery
Neural scaling laws for phenotypic drug discovery Open
Recent breakthroughs by deep neural networks (DNNs) in natural language processing (NLP) and computer vision have been driven by a scale-up of models and data rather than the discovery of novel computing paradigms. Here, we investigate if …
View article: Fixing the problems of deep neural networks will require better training data and learning algorithms
Fixing the problems of deep neural networks will require better training data and learning algorithms Open
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. We show that this problem is worsening as DNNs …
View article: Diagnosing and exploiting the computational demands of videos games for deep reinforcement learning
Diagnosing and exploiting the computational demands of videos games for deep reinforcement learning Open
Humans learn by interacting with their environments and perceiving the outcomes of their actions. A landmark in artificial intelligence has been the development of deep reinforcement learning (dRL) algorithms capable of doing the same in v…
View article: Harmonizing the visual strategies of image-computable models with humans yields more performant and interpretable models of primate visual system function.
Harmonizing the visual strategies of image-computable models with humans yields more performant and interpretable models of primate visual system function. Open
International audience
View article: Unlocking Feature Visualization for Deeper Networks with MAgnitude Constrained Optimization
Unlocking Feature Visualization for Deeper Networks with MAgnitude Constrained Optimization Open
Feature visualization has gained substantial popularity, particularly after the influential work by Olah et al. in 2017, which established it as a crucial tool for explainability. However, its widespread adoption has been limited due to a …
View article: Performance-optimized deep neural networks are evolving into worse models of inferotemporal visual cortex
Performance-optimized deep neural networks are evolving into worse models of inferotemporal visual cortex Open
One of the most impactful findings in computational neuroscience over the past decade is that the object recognition accuracy of deep neural networks (DNNs) correlates with their ability to predict neural responses to natural images in the…
View article: Adversarial alignment: Breaking the trade-off between the strength of an attack and its relevance to human perception
Adversarial alignment: Breaking the trade-off between the strength of an attack and its relevance to human perception Open
Deep neural networks (DNNs) are known to have a fundamental sensitivity to adversarial attacks, perturbations of the input that are imperceptible to humans yet powerful enough to change the visual decision of a model. Adversarial attacks h…
View article: Fixing the problems of deep neural networks will require better training data and learning algorithms
Fixing the problems of deep neural networks will require better training data and learning algorithms Open
Bowers et al. argue that deep neural networks (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. We show that this problem is wor…
View article: Toward modeling visual routines of object segmentation with biologically inspired recurrent vision models
Toward modeling visual routines of object segmentation with biologically inspired recurrent vision models Open
International audience
View article: Harmonizing the object recognition strategies of deep neural networks with humans.
Harmonizing the object recognition strategies of deep neural networks with humans. Open
The many successes of deep neural networks (DNNs) over the past decade have largely been driven by computational scale rather than insights from biological intelligence. Here, we explore if these trends have also carried concomitant improv…
View article: Harmonizing the object recognition strategies of deep neural networks with humans
Harmonizing the object recognition strategies of deep neural networks with humans Open
The many successes of deep neural networks (DNNs) over the past decade have largely been driven by computational scale rather than insights from biological intelligence. Here, we explore if these trends have also carried concomitant improv…
View article: Fluorescently labeled nuclear morphology is highly informative of neurotoxicity
Fluorescently labeled nuclear morphology is highly informative of neurotoxicity Open
Neurotoxicity can be detected in live microscopy by morphological changes such as retraction of neurites, fragmentation, blebbing of the neuronal soma and ultimately the disappearance of fluorescently labeled neurons. However, quantificati…
View article: Deceptive learning in histopathology
Deceptive learning in histopathology Open
Deep learning holds immense potential for histopathology, automating tasks that are simple for expert pathologists, and revealing novel biology for tasks that were previously considered difficult or impossible to solve by eye alone. Howeve…
View article: Understanding the Computational Demands Underlying Visual Reasoning
Understanding the Computational Demands Underlying Visual Reasoning Open
Visual understanding requires comprehending complex visual relations between objects within a scene. Here, we seek to characterize the computational demands for abstract visual reasoning. We do this by systematically assessing the ability …
View article: How and What to Learn: Taxonomizing Self-Supervised Learning for 3D Action Recognition
How and What to Learn: Taxonomizing Self-Supervised Learning for 3D Action Recognition Open
International audience
View article: Superhuman cell death detection with biomarker-optimized neural networks
Superhuman cell death detection with biomarker-optimized neural networks Open
High-throughput microscopy has outpaced analysis; biomarker-optimized CNNs are a generalizable, fast, and interpretable solution.
View article: Tracking Without Re-recognition in Humans and Machines.
Tracking Without Re-recognition in Humans and Machines. Open
Imagine trying to track one particular fruitfly in a swarm of hundreds. Higher biological visual systems have evolved to track moving objects by relying on both appearance and motion features. We investigate if state-of-the-art deep neural…
View article: The Challenge of Appearance-Free Object Tracking with Feedforward Neural Networks
The Challenge of Appearance-Free Object Tracking with Feedforward Neural Networks Open
Nearly all models for object tracking with artificial neural networks depend on appearance features extracted from a "backbone" architecture, designed for object recognition. Indeed, significant progress on object tracking has been spurred…
View article: The Challenge of Appearance-Free Object Tracking with Feedforward Neural\n Networks
The Challenge of Appearance-Free Object Tracking with Feedforward Neural\n Networks Open
Nearly all models for object tracking with artificial neural networks depend\non appearance features extracted from a "backbone" architecture, designed for\nobject recognition. Indeed, significant progress on object tracking has been\nspur…
View article: Uncovering the circuit mechanisms that shape contextual phenomena with task-optimized recurrent neural networks
Uncovering the circuit mechanisms that shape contextual phenomena with task-optimized recurrent neural networks Open
Neurons in visual cortex are sensitive to context. Neural responses to stimuli presented within their classical receptive fields (CRFs) are modulated by the presence of other stimuli – in their CRF and their surrounding extra-classical rec…
View article: Understanding the computational demands underlying visual reasoning
Understanding the computational demands underlying visual reasoning Open
Visual understanding requires comprehending complex visual relations between objects within a scene. Here, we seek to characterize the computational demands for abstract visual reasoning. We do this by systematically assessing the ability …
View article: Tracking Without Re-recognition in Humans and Machines
Tracking Without Re-recognition in Humans and Machines Open
Imagine trying to track one particular fruitfly in a swarm of hundreds. Higher biological visual systems have evolved to track moving objects by relying on both appearance and motion features. We investigate if state-of-the-art deep neural…
View article: Recurrent neural circuits for contour detection
Recurrent neural circuits for contour detection Open
We introduce a deep recurrent neural network architecture that approximates visual cortical circuits. We show that this architecture, which we refer to as the gamma-net, learns to solve contour detection tasks with better sample efficiency…