Martin Wattenberg
Research Subjects
Martin Wattenberg has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Fernanda Viegas
Google Brain Cambridge MA
4 shared publications - Daniel Smilkov
Palo Alto CA USA
2 shared publications - Emily Reif
Google research
2 shared publications - Adam Pearce
Yannick Assogba Adam Pearce and Madison Elliot are with Google Research
1 shared publication - Andy Coenen
Google research
1 shared publication - Ann Yuan
Google research
1 shared publication
Latest Publications
- Open Access
- 9 April 2021
- Published by Springer Nature in Computational Brain & Behavior
- Vol. 4 (3) , 251-263
Neural Networks Trained on Natural Scenes Exhibit Gestalt Closure
The Gestalt laws of perceptual organization, which describe how visual elements in an image are grouped and interpreted, have traditionally been thought of as innate. Given past research showing that these laws have ecological validity, we investigate whether deep learning methods infer Gestalt laws from the statistics of natural scenes. We examine the law of closure, which asserts that human visual perception tends to “close the gap” by assembling elements that can jointly be interpreted as a complete figure or object. We demonstrate that a state-of-the-art convolutional neural network, trained to classify natural images, exhibits closure on synthetic displays of edge fragments, as assessed by similarity of internal representations. This finding provides further support for the hypothesis that the human perceptual system is even more elegant than the Gestaltists imagined: a single law—adaptation to the statistical structure of the environment—might suffice as fundamental.
- 9 April 2021
- Published by Springer Nature in Computational Brain & Behavior
- Vol. 4 (3) , 251-263
Open AccessNeural Networks Trained on Natural Scenes Exhibit Gestalt Closure
The Gestalt laws of perceptual organization, which describe how visual elements in an image are grouped and interpreted, have traditionally been thought of as innate. Given past research showing that these laws have ecological validity, we investigate whether deep learning methods infer Gestalt laws from the statistics of natural scenes. We examine the law of closure, which asserts that human visual perception tends to “close the gap” by assembling elements that can jointly be interpreted as a complete figure or object. We demonstrate that a state-of-the-art convolutional neural network, trained to classify natural images, exhibits closure on synthetic displays of edge fragments, as assessed by similarity of internal representations. This finding provides further support for the hypothesis that the human perceptual system is even more elegant than the Gestaltists imagined: a single law—adaptation to the statistical structure of the environment—might suffice as fundamental.
- Open Access
Visualizing and Measuring the Geometry of BERT
Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks, these networks appear to extract generally useful linguistic features. A natural question is how such networks represent this information internally. This paper describes qualitative and quantitative investigations of one particularly effective model, BERT. At a high level, linguistic features seem to be represented in separate semantic and syntactic subspaces. We find evidence of a fine-grained geometric representation of word senses. We also present empirical descriptions of syntactic representations in both attention matrices and individual word embeddings, as well as a mathematical argument to explain the geometry of these representations.
Open AccessVisualizing and Measuring the Geometry of BERT
Transformer architectures show significant promise for natural language processing. Given that a single pretrained model can be fine-tuned to perform well on many different tasks, these networks appear to extract generally useful linguistic features. A natural question is how such networks represent this information internally. This paper describes qualitative and quantitative investigations of one particularly effective model, BERT. At a high level, linguistic features seem to be represented in separate semantic and syntactic subspaces. We find evidence of a fine-grained geometric representation of word senses. We also present empirical descriptions of syntactic representations in both attention matrices and individual word embeddings, as well as a mathematical argument to explain the geometry of these representations.
- Top Cited PapersOpen Access
- 20 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. PP (10772626) , 1
The What-If Tool: Interactive Probing of Machine Learning Models
A key challenge in developing and deploying Machine Learning (ML) systems is understanding their performance across a wide range of inputs. To address this challenge, we created the What-If Tool, an open-source application that allows practitioners to probe, visualize, and analyze ML systems, with minimal coding. The What-If Tool lets practitioners test performance in hypothetical situations, analyze the importance of different data features, and visualize model behavior across multiple models and subsets of input data. It also lets practitioners measure systems according to multiple ML fairness metrics. We describe the design of the tool, and report on real-life usage at different organizations.
- 20 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Visualization and Computer Graphics
- Vol. PP (10772626) , 1
Top Cited PapersOpen AccessThe What-If Tool: Interactive Probing of Machine Learning Models
A key challenge in developing and deploying Machine Learning (ML) systems is understanding their performance across a wide range of inputs. To address this challenge, we created the What-If Tool, an open-source application that allows practitioners to probe, visualize, and analyze ML systems, with minimal coding. The What-If Tool lets practitioners test performance in hypothetical situations, analyze the importance of different data features, and visualize model behavior across multiple models and subsets of input data. It also lets practitioners measure systems according to multiple ML fairness metrics. We describe the design of the tool, and report on real-life usage at different organizations.
- Open Access
Adversarial Spheres
State of the art computer vision models have been shown to be vulnerable to small adversarial perturbations of the input. In other words, most images in the data distribution are both correctly classified by the model and are very close to a visually similar misclassified image. Despite substantial research interest, the cause of the phenomenon is still poorly understood and remains unsolved. We hypothesize that this counter intuitive behavior is a naturally occurring result of the high dimensional geometry of the data manifold. As a first step towards exploring this hypothesis, we study a simple synthetic dataset of classifying between two concentric high dimensional spheres. For this dataset we show a fundamental tradeoff between the amount of test error and the average distance to nearest error. In particular, we prove that any model which misclassifies a small constant fraction of a sphere will be vulnerable to adversarial perturbations of size $O(1/\sqrt{d})$. Surprisingly, when we train several different architectures on this dataset, all of their error sets naturally approach this theoretical bound. As a result of the theory, the vulnerability of neural networks to small adversarial perturbations is a logical consequence of the amount of test error observed. We hope that our theoretical analysis of this very simple case will point the way forward to explore how the geometry of complex real-world data sets leads to adversarial examples.Open AccessAdversarial Spheres
State of the art computer vision models have been shown to be vulnerable to small adversarial perturbations of the input. In other words, most images in the data distribution are both correctly classified by the model and are very close to a visually similar misclassified image. Despite substantial research interest, the cause of the phenomenon is still poorly understood and remains unsolved. We hypothesize that this counter intuitive behavior is a naturally occurring result of the high dimensional geometry of the data manifold. As a first step towards exploring this hypothesis, we study a simple synthetic dataset of classifying between two concentric high dimensional spheres. For this dataset we show a fundamental tradeoff between the amount of test error and the average distance to nearest error. In particular, we prove that any model which misclassifies a small constant fraction of a sphere will be vulnerable to adversarial perturbations of size $O(1/\sqrt{d})$. Surprisingly, when we train several different architectures on this dataset, all of their error sets naturally approach this theoretical bound. As a result of the theory, the vulnerability of neural networks to small adversarial perturbations is a logical consequence of the amount of test error observed. We hope that our theoretical analysis of this very simple case will point the way forward to explore how the geometry of complex real-world data sets leads to adversarial examples. - Open Access
Direct-Manipulation Visualization of Deep Networks
The recent successes of deep learning have led to a wave of interest from non-experts. Gaining an understanding of this technology, however, is difficult. While the theory is important, it is also helpful for novices to develop an intuitive feel for the effect of different hyperparameters and structural variations. We describe TensorFlow Playground, an interactive, open sourced visualization that allows users to experiment via direct manipulation rather than coding, enabling them to quickly build an intuition about neural nets.
Open AccessDirect-Manipulation Visualization of Deep Networks
The recent successes of deep learning have led to a wave of interest from non-experts. Gaining an understanding of this technology, however, is difficult. While the theory is important, it is also helpful for novices to develop an intuitive feel for the effect of different hyperparameters and structural variations. We describe TensorFlow Playground, an interactive, open sourced visualization that allows users to experiment via direct manipulation rather than coding, enabling them to quickly build an intuition about neural nets.
- Open Access
Embedding Projector: Interactive Visualization and Interpretation of Embeddings
Embeddings are ubiquitous in machine learning, appearing in recommender systems, NLP, and many other applications. Researchers and developers often need to explore the properties of a specific embedding, and one way to analyze embeddings is to visualize them. We present the Embedding Projector, a tool for interactive visualization and interpretation of embeddings.
Open AccessEmbedding Projector: Interactive Visualization and Interpretation of Embeddings
Embeddings are ubiquitous in machine learning, appearing in recommender systems, NLP, and many other applications. Researchers and developers often need to explore the properties of a specific embedding, and one way to analyze embeddings is to visualize them. We present the Embedding Projector, a tool for interactive visualization and interpretation of embeddings.