**Click to Read Paper**

Batched Gaussian Process Bandit Optimization via Determinantal Point Processes

Nov 13, 2016

Tarun Kathuria, Amit Deshpande, Pushmeet Kohli

Nov 13, 2016

Tarun Kathuria, Amit Deshpande, Pushmeet Kohli

* To appear at NIPS 2016

**Click to Read Paper**

Efficient Energy Minimization for Enforcing Statistics

Jul 30, 2013

Yongsub Lim, Kyomin Jung, Pushmeet Kohli

Jul 30, 2013

Yongsub Lim, Kyomin Jung, Pushmeet Kohli

**Click to Read Paper**

Multi-dimensional Parametric Mincuts for Constrained MAP Inference

Jul 30, 2013

Yongsub Lim, Kyomin Jung, Pushmeet Kohli

Jul 30, 2013

Yongsub Lim, Kyomin Jung, Pushmeet Kohli

* 19 pages

**Click to Read Paper**

Curvature Prior for MRF-based Segmentation and Shape Inpainting

Sep 07, 2011

Alexander Shekhovtsov, Pushmeet Kohli, Carsten Rother

Most image labeling problems such as segmentation and image reconstruction are fundamentally ill-posed and suffer from ambiguities and noise. Higher order image priors encode high level structural dependencies between pixels and are key to overcoming these problems. However, these priors in general lead to computationally intractable models. This paper addresses the problem of discovering compact representations of higher order priors which allow efficient inference. We propose a framework for solving this problem which uses a recently proposed representation of higher order functions where they are encoded as lower envelopes of linear functions. Maximum a Posterior inference on our learned models reduces to minimizing a pairwise function of discrete variables, which can be done approximately using standard methods. Although this is a primarily theoretical paper, we also demonstrate the practical effectiveness of our framework on the problem of learning a shape prior for image segmentation and reconstruction. We show that our framework can learn a compact representation that approximates a prior that encourages low curvature shapes. We evaluate the approximation accuracy, discuss properties of the trained model, and show various results for shape inpainting and image segmentation.
Sep 07, 2011

Alexander Shekhovtsov, Pushmeet Kohli, Carsten Rother

* 17 pages, 16 figures

**Click to Read Paper**

Learning an Interactive Segmentation System

Dec 13, 2009

Hannes Nickisch, Pushmeet Kohli, Carsten Rother

Dec 13, 2009

Hannes Nickisch, Pushmeet Kohli, Carsten Rother

* 11 pages, 7 figures, 4 tables

**Click to Read Paper**

Deep disentangled representations for volumetric reconstruction

Oct 12, 2016

Edward Grant, Pushmeet Kohli, Marcel van Gerven

Oct 12, 2016

Edward Grant, Pushmeet Kohli, Marcel van Gerven

**Click to Read Paper**

Verification of deep probabilistic models

Dec 06, 2018

Krishnamurthy Dvijotham, Marta Garnelo, Alhussein Fawzi, Pushmeet Kohli

Probabilistic models are a critical part of the modern deep learning toolbox - ranging from generative models (VAEs, GANs), sequence to sequence models used in machine translation and speech processing to models over functional spaces (conditional neural processes, neural processes). Given the size and complexity of these models, safely deploying them in applications requires the development of tools to analyze their behavior rigorously and provide some guarantees that these models are consistent with a list of desirable properties or specifications. For example, a machine translation model should produce semantically equivalent outputs for innocuous changes in the input to the model. A functional regression model that is learning a distribution over monotonic functions should predict a larger value at a larger input. Verification of these properties requires a new framework that goes beyond notions of verification studied in deterministic feedforward networks, since requiring worst-case guarantees in probabilistic models is likely to produce conservative or vacuous results. We propose a novel formulation of verification for deep probabilistic models that take in conditioning inputs and sample latent variables in the course of producing an output: We require that the output of the model satisfies a linear constraint with high probability over the sampling of latent variables and for every choice of conditioning input to the model. We show that rigorous lower bounds on the probability that the constraint is satisfied can be obtained efficiently. Experiments with neural processes show that several properties of interest while modeling functional spaces can be modeled within this framework (monotonicity, convexity) and verified efficiently using our algorithms
Dec 06, 2018

Krishnamurthy Dvijotham, Marta Garnelo, Alhussein Fawzi, Pushmeet Kohli

* Accepted to NeurIPS 2018 Workshop on Security in Machine Learning

**Click to Read Paper**

Batched Large-scale Bayesian Optimization in High-dimensional Spaces

May 16, 2018

Zi Wang, Clement Gehring, Pushmeet Kohli, Stefanie Jegelka

May 16, 2018

Zi Wang, Clement Gehring, Pushmeet Kohli, Stefanie Jegelka

* Proceedings of the 21st International Conference on Artificial Intelligence and Statistics (AISTATS) 2018, Lanzarote, Spain

**Click to Read Paper**

Batched High-dimensional Bayesian Optimization via Structural Kernel Learning

Jan 06, 2018

Zi Wang, Chengtao Li, Stefanie Jegelka, Pushmeet Kohli

Jan 06, 2018

Zi Wang, Chengtao Li, Stefanie Jegelka, Pushmeet Kohli

* Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia, PMLR 70, 2017

**Click to Read Paper**

Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learning

Nov 07, 2017

Junhyuk Oh, Satinder Singh, Honglak Lee, Pushmeet Kohli

Nov 07, 2017

Junhyuk Oh, Satinder Singh, Honglak Lee, Pushmeet Kohli

* ICML 2017

**Click to Read Paper**

Semantic Code Repair using Neuro-Symbolic Transformation Networks

Oct 30, 2017

Jacob Devlin, Jonathan Uesato, Rishabh Singh, Pushmeet Kohli

We study the problem of semantic code repair, which can be broadly defined as automatically fixing non-syntactic bugs in source code. The majority of past work in semantic code repair assumed access to unit tests against which candidate repairs could be validated. In contrast, the goal here is to develop a strong statistical model to accurately predict both bug locations and exact fixes without access to information about the intended correct behavior of the program. Achieving such a goal requires a robust contextual repair model, which we train on a large corpus of real-world source code that has been augmented with synthetically injected bugs. Our framework adopts a two-stage approach where first a large set of repair candidates are generated by rule-based processors, and then these candidates are scored by a statistical model using a novel neural network architecture which we refer to as Share, Specialize, and Compete. Specifically, the architecture (1) generates a shared encoding of the source code using an RNN over the abstract syntax tree, (2) scores each candidate repair using specialized network modules, and (3) then normalizes these scores together so they can compete against one another in comparable probability space. We evaluate our model on a real-world test set gathered from GitHub containing four common categories of bugs. Our model is able to predict the exact correct repair 41\% of the time with a single guess, compared to 13\% accuracy for an attentional sequence-to-sequence model.
Oct 30, 2017

Jacob Devlin, Jonathan Uesato, Rishabh Singh, Pushmeet Kohli

**Click to Read Paper**

Learning Continuous Semantic Representations of Symbolic Expressions

Jun 10, 2017

Miltiadis Allamanis, Pankajan Chanthirasegaran, Pushmeet Kohli, Charles Sutton

Jun 10, 2017

Miltiadis Allamanis, Pankajan Chanthirasegaran, Pushmeet Kohli, Charles Sutton

* Accepted to ICML 2017

**Click to Read Paper**

**Click to Read Paper**

Deep Multi-Modal Image Correspondence Learning

Dec 05, 2016

Chen Liu, Jiajun Wu, Pushmeet Kohli, Yasutaka Furukawa

Dec 05, 2016

Chen Liu, Jiajun Wu, Pushmeet Kohli, Yasutaka Furukawa

**Click to Read Paper**

PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

Oct 16, 2016

Michael Figurnov, Aijan Ibraimova, Dmitry Vetrov, Pushmeet Kohli

Oct 16, 2016

Michael Figurnov, Aijan Ibraimova, Dmitry Vetrov, Pushmeet Kohli

* NIPS 2016

**Click to Read Paper**

Time-Sensitive Bayesian Information Aggregation for Crowdsourcing Systems

Apr 18, 2016

Matteo Venanzi, John Guiver, Pushmeet Kohli, Nick Jennings

Apr 18, 2016

Matteo Venanzi, John Guiver, Pushmeet Kohli, Nick Jennings

**Click to Read Paper**

* Short version of this paper will appear in HCOMP'15

**Click to Read Paper**

Multi-utility Learning: Structured-output Learning with Multiple Annotation-specific Loss Functions

Jun 23, 2014

Roman Shapovalov, Dmitry Vetrov, Anton Osokin, Pushmeet Kohli

Jun 23, 2014

Roman Shapovalov, Dmitry Vetrov, Anton Osokin, Pushmeet Kohli

**Click to Read Paper**

Partition-Merge: Distributed Inference and Modularity Optimization

Sep 24, 2013

Vincent Blondel, Kyomin Jung, Pushmeet Kohli, Devavrat Shah

Sep 24, 2013

Vincent Blondel, Kyomin Jung, Pushmeet Kohli, Devavrat Shah

* 19 pages, 1 figure

**Click to Read Paper**