Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Pushmeet Kohli

Pushmeet Kohli

Challenges in Detoxifying Language Models


Sep 15, 2021
Johannes Welbl, Amelia Glaese, Jonathan Uesato, Sumanth Dathathri, John Mellor, Lisa Anne Hendricks, Kirsty Anderson, Pushmeet Kohli, Ben Coppin, Po-Sen Huang

Figure 1 for Challenges in Detoxifying Language Models
Figure 2 for Challenges in Detoxifying Language Models
Figure 3 for Challenges in Detoxifying Language Models
Figure 4 for Challenges in Detoxifying Language Models
* 23 pages, 6 figures, published in Findings of EMNLP 2021 
  Access Paper or Ask Questions

Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs


Jun 26, 2021
Dan Rosenbaum, Marta Garnelo, Michal Zielinski, Charlie Beattie, Ellen Clancy, Andrea Huber, Pushmeet Kohli, Andrew W. Senior, John Jumper, Carl Doersch, S. M. Ali Eslami, Olaf Ronneberger, Jonas Adler

Figure 1 for Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs
Figure 2 for Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs
Figure 3 for Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs
Figure 4 for Inferring a Continuous Distribution of Atom Coordinates from Cryo-EM Images using VAEs

  Access Paper or Ask Questions

Improved Branch and Bound for Neural Network Verification via Lagrangian Decomposition


Apr 14, 2021
Alessandro De Palma, Rudy Bunel, Alban Desmaison, Krishnamurthy Dvijotham, Pushmeet Kohli, Philip H. S. Torr, M. Pawan Kumar

Figure 1 for Improved Branch and Bound for Neural Network Verification via Lagrangian Decomposition
Figure 2 for Improved Branch and Bound for Neural Network Verification via Lagrangian Decomposition
Figure 3 for Improved Branch and Bound for Neural Network Verification via Lagrangian Decomposition
Figure 4 for Improved Branch and Bound for Neural Network Verification via Lagrangian Decomposition
* Submitted for review to JMLR. This is an extended version of our paper in the UAI-20 conference (arXiv:2002.10410) 
  Access Paper or Ask Questions

Solving Mixed Integer Programs Using Neural Networks


Dec 23, 2020
Vinod Nair, Sergey Bartunov, Felix Gimeno, Ingrid von Glehn, Pawel Lichocki, Ivan Lobov, Brendan O'Donoghue, Nicolas Sonnerat, Christian Tjandraatmadja, Pengming Wang, Ravichandra Addanki, Tharindi Hapuarachchi, Thomas Keck, James Keeling, Pushmeet Kohli, Ira Ktena, Yujia Li, Oriol Vinyals, Yori Zwols

Figure 1 for Solving Mixed Integer Programs Using Neural Networks
Figure 2 for Solving Mixed Integer Programs Using Neural Networks
Figure 3 for Solving Mixed Integer Programs Using Neural Networks
Figure 4 for Solving Mixed Integer Programs Using Neural Networks

  Access Paper or Ask Questions

Autoencoding Variational Autoencoder


Dec 07, 2020
A. Taylan Cemgil, Sumedh Ghaisas, Krishnamurthy Dvijotham, Sven Gowal, Pushmeet Kohli

Figure 1 for Autoencoding Variational Autoencoder
Figure 2 for Autoencoding Variational Autoencoder
Figure 3 for Autoencoding Variational Autoencoder
Figure 4 for Autoencoding Variational Autoencoder
* Neurips 2020 
  Access Paper or Ask Questions

Towards transformation-resilient provenance detection of digital media


Nov 14, 2020
Jamie Hayes, Krishnamurthy, Dvijotham, Yutian Chen, Sander Dieleman, Pushmeet Kohli, Norman Casagrande

Figure 1 for Towards transformation-resilient provenance detection of digital media
Figure 2 for Towards transformation-resilient provenance detection of digital media
Figure 3 for Towards transformation-resilient provenance detection of digital media
Figure 4 for Towards transformation-resilient provenance detection of digital media

  Access Paper or Ask Questions

Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming


Nov 03, 2020
Sumanth Dathathri, Krishnamurthy Dvijotham, Alexey Kurakin, Aditi Raghunathan, Jonathan Uesato, Rudy Bunel, Shreya Shankar, Jacob Steinhardt, Ian Goodfellow, Percy Liang, Pushmeet Kohli

Figure 1 for Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming
Figure 2 for Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming
Figure 3 for Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming
Figure 4 for Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming

  Access Paper or Ask Questions

Training Generative Adversarial Networks by Solving Ordinary Differential Equations


Oct 28, 2020
Chongli Qin, Yan Wu, Jost Tobias Springenberg, Andrew Brock, Jeff Donahue, Timothy P. Lillicrap, Pushmeet Kohli

Figure 1 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations
Figure 2 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations
Figure 3 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations
Figure 4 for Training Generative Adversarial Networks by Solving Ordinary Differential Equations

  Access Paper or Ask Questions

Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples


Oct 27, 2020
Sven Gowal, Chongli Qin, Jonathan Uesato, Timothy Mann, Pushmeet Kohli

Figure 1 for Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples
Figure 2 for Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples
Figure 3 for Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples
Figure 4 for Uncovering the Limits of Adversarial Training against Norm-Bounded Adversarial Examples

  Access Paper or Ask Questions

Contrastive Training for Improved Out-of-Distribution Detection


Jul 10, 2020
Jim Winkens, Rudy Bunel, Abhijit Guha Roy, Robert Stanforth, Vivek Natarajan, Joseph R. Ledsam, Patricia MacWilliams, Pushmeet Kohli, Alan Karthikesalingam, Simon Kohl, Taylan Cemgil, S. M. Ali Eslami, Olaf Ronneberger

Figure 1 for Contrastive Training for Improved Out-of-Distribution Detection
Figure 2 for Contrastive Training for Improved Out-of-Distribution Detection
Figure 3 for Contrastive Training for Improved Out-of-Distribution Detection
Figure 4 for Contrastive Training for Improved Out-of-Distribution Detection

  Access Paper or Ask Questions

Evaluating the Apperception Engine


Jul 09, 2020
Richard Evans, Jose Hernandez-Orallo, Johannes Welbl, Pushmeet Kohli, Marek Sergot

Figure 1 for Evaluating the Apperception Engine
Figure 2 for Evaluating the Apperception Engine
Figure 3 for Evaluating the Apperception Engine
Figure 4 for Evaluating the Apperception Engine
* arXiv admin note: substantial text overlap with arXiv:1910.02227 
  Access Paper or Ask Questions

Strong Generalization and Efficiency in Neural Programs


Jul 08, 2020
Yujia Li, Felix Gimeno, Pushmeet Kohli, Oriol Vinyals

Figure 1 for Strong Generalization and Efficiency in Neural Programs
Figure 2 for Strong Generalization and Efficiency in Neural Programs
Figure 3 for Strong Generalization and Efficiency in Neural Programs
Figure 4 for Strong Generalization and Efficiency in Neural Programs

  Access Paper or Ask Questions

Lagrangian Decomposition for Neural Network Verification


Feb 24, 2020
Rudy Bunel, Alessandro De Palma, Alban Desmaison, Krishnamurthy Dvijotham, Pushmeet Kohli, Philip H. S. Torr, M. Pawan Kumar

Figure 1 for Lagrangian Decomposition for Neural Network Verification
Figure 2 for Lagrangian Decomposition for Neural Network Verification
Figure 3 for Lagrangian Decomposition for Neural Network Verification
Figure 4 for Lagrangian Decomposition for Neural Network Verification

  Access Paper or Ask Questions

Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations


Dec 06, 2019
Sven Gowal, Chongli Qin, Po-Sen Huang, Taylan Cemgil, Krishnamurthy Dvijotham, Timothy Mann, Pushmeet Kohli

Figure 1 for Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations
Figure 2 for Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations
Figure 3 for Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations
Figure 4 for Achieving Robustness in the Wild via Adversarial Mixing with Disentangled Representations

  Access Paper or Ask Questions

Reducing Sentiment Bias in Language Models via Counterfactual Evaluation


Nov 08, 2019
Po-Sen Huang, Huan Zhang, Ray Jiang, Robert Stanforth, Johannes Welbl, Jack Rae, Vishal Maini, Dani Yogatama, Pushmeet Kohli

Figure 1 for Reducing Sentiment Bias in Language Models via Counterfactual Evaluation
Figure 2 for Reducing Sentiment Bias in Language Models via Counterfactual Evaluation
Figure 3 for Reducing Sentiment Bias in Language Models via Counterfactual Evaluation
Figure 4 for Reducing Sentiment Bias in Language Models via Counterfactual Evaluation

  Access Paper or Ask Questions

Learning Transferable Graph Exploration


Oct 28, 2019
Hanjun Dai, Yujia Li, Chenglong Wang, Rishabh Singh, Po-Sen Huang, Pushmeet Kohli

Figure 1 for Learning Transferable Graph Exploration
Figure 2 for Learning Transferable Graph Exploration
Figure 3 for Learning Transferable Graph Exploration
Figure 4 for Learning Transferable Graph Exploration
* To appear in NeurIPS 2019 
  Access Paper or Ask Questions

An Alternative Surrogate Loss for PGD-based Adversarial Testing


Oct 21, 2019
Sven Gowal, Jonathan Uesato, Chongli Qin, Po-Sen Huang, Timothy Mann, Pushmeet Kohli

Figure 1 for An Alternative Surrogate Loss for PGD-based Adversarial Testing
Figure 2 for An Alternative Surrogate Loss for PGD-based Adversarial Testing
Figure 3 for An Alternative Surrogate Loss for PGD-based Adversarial Testing
Figure 4 for An Alternative Surrogate Loss for PGD-based Adversarial Testing

  Access Paper or Ask Questions

Making sense of sensory input


Oct 05, 2019
Richard Evans, Jose Hernandez-Orallo, Johannes Welbl, Pushmeet Kohli, Marek Sergot

Figure 1 for Making sense of sensory input
Figure 2 for Making sense of sensory input
Figure 3 for Making sense of sensory input
Figure 4 for Making sense of sensory input

  Access Paper or Ask Questions

CLEVRER: CoLlision Events for Video REpresentation and Reasoning


Oct 03, 2019
Kexin Yi, Chuang Gan, Yunzhu Li, Pushmeet Kohli, Jiajun Wu, Antonio Torralba, Joshua B. Tenenbaum

Figure 1 for CLEVRER: CoLlision Events for Video REpresentation and Reasoning
Figure 2 for CLEVRER: CoLlision Events for Video REpresentation and Reasoning
Figure 3 for CLEVRER: CoLlision Events for Video REpresentation and Reasoning
Figure 4 for CLEVRER: CoLlision Events for Video REpresentation and Reasoning
* The first two authors contributed equally to this work. Project page: http://clevrer.csail.mit.edu/ 
  Access Paper or Ask Questions

Branch and Bound for Piecewise Linear Neural Network Verification


Sep 14, 2019
Rudy Bunel, Jingyue Lu, Ilker Turkaslan, Philip H. S. Torr, Pushmeet Kohli, M. Pawan Kumar

Figure 1 for Branch and Bound for Piecewise Linear Neural Network Verification
Figure 2 for Branch and Bound for Piecewise Linear Neural Network Verification
Figure 3 for Branch and Bound for Piecewise Linear Neural Network Verification
Figure 4 for Branch and Bound for Piecewise Linear Neural Network Verification

  Access Paper or Ask Questions

Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation


Sep 03, 2019
Po-Sen Huang, Robert Stanforth, Johannes Welbl, Chris Dyer, Dani Yogatama, Sven Gowal, Krishnamurthy Dvijotham, Pushmeet Kohli

Figure 1 for Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation
Figure 2 for Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation
Figure 3 for Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation
Figure 4 for Achieving Verified Robustness to Symbol Substitutions via Interval Bound Propagation
* Accepted in EMNLP 2019 
  Access Paper or Ask Questions

Adversarial Robustness through Local Linearization


Jul 04, 2019
Chongli Qin, James Martens, Sven Gowal, Dilip Krishnan, Krishnamurthy, Dvijotham, Alhussein Fawzi, Soham De, Robert Stanforth, Pushmeet Kohli

Figure 1 for Adversarial Robustness through Local Linearization
Figure 2 for Adversarial Robustness through Local Linearization
Figure 3 for Adversarial Robustness through Local Linearization
Figure 4 for Adversarial Robustness through Local Linearization

  Access Paper or Ask Questions

Are Labels Required for Improving Adversarial Robustness?


May 31, 2019
Jonathan Uesato*, Jean-Baptiste Alayrac*, Po-Sen Huang*, Robert Stanforth, Alhussein Fawzi, Pushmeet Kohli

Figure 1 for Are Labels Required for Improving Adversarial Robustness?
Figure 2 for Are Labels Required for Improving Adversarial Robustness?
Figure 3 for Are Labels Required for Improving Adversarial Robustness?
Figure 4 for Are Labels Required for Improving Adversarial Robustness?

  Access Paper or Ask Questions

A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities


May 30, 2019
Simon A. A. Kohl, Bernardino Romera-Paredes, Klaus H. Maier-Hein, Danilo Jimenez Rezende, S. M. Ali Eslami, Pushmeet Kohli, Andrew Zisserman, Olaf Ronneberger

Figure 1 for A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
Figure 2 for A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
Figure 3 for A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
Figure 4 for A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
* 25 pages, 15 figures 
  Access Paper or Ask Questions

REGAL: Transfer Learning For Fast Optimization of Computation Graphs


May 30, 2019
Aditya Paliwal, Felix Gimeno, Vinod Nair, Yujia Li, Miles Lubin, Pushmeet Kohli, Oriol Vinyals

Figure 1 for REGAL: Transfer Learning For Fast Optimization of Computation Graphs
Figure 2 for REGAL: Transfer Learning For Fast Optimization of Computation Graphs
Figure 3 for REGAL: Transfer Learning For Fast Optimization of Computation Graphs
Figure 4 for REGAL: Transfer Learning For Fast Optimization of Computation Graphs

  Access Paper or Ask Questions

Structured agents for physical construction


May 13, 2019
Victor Bapst, Alvaro Sanchez-Gonzalez, Carl Doersch, Kimberly L. Stachenfeld, Pushmeet Kohli, Peter W. Battaglia, Jessica B. Hamrick

Figure 1 for Structured agents for physical construction
Figure 2 for Structured agents for physical construction
Figure 3 for Structured agents for physical construction
Figure 4 for Structured agents for physical construction
* ICML 2019 
  Access Paper or Ask Questions

Graph Matching Networks for Learning the Similarity of Graph Structured Objects


May 12, 2019
Yujia Li, Chenjie Gu, Thomas Dullien, Oriol Vinyals, Pushmeet Kohli

Figure 1 for Graph Matching Networks for Learning the Similarity of Graph Structured Objects
Figure 2 for Graph Matching Networks for Learning the Similarity of Graph Structured Objects
Figure 3 for Graph Matching Networks for Learning the Similarity of Graph Structured Objects
Figure 4 for Graph Matching Networks for Learning the Similarity of Graph Structured Objects
* Accepted as a conference paper at ICML 2019 
  Access Paper or Ask Questions