Pushmeet Kohli

42

Papers

788

Total Citations

Papers (42)

Learning Disentangled Representations with Semi-Supervised Deep Generative Models

NeurIPS 2017arXiv

PerforatedCNNs: Acceleration through Elimination of Redundant Convolutions

NeurIPS 2016arXiv

Batched Gaussian Process Bandit Optimization via Determinantal Point Processes

NeurIPS 2016arXiv

Neural Program Meta-Induction

NeurIPS 2017arXiv

Adaptive Neural Compilation

NeurIPS 2016arXiv

Scaling Wearable Foundation Models

SensorLM: Learning the Language of Wearable Sensors

NeurIPS 2025arXiv

Knowing When to Stop: Evaluation and Verification of Conformity to Output-Size Specifications

Achieving Robustness in the Wild via Adversarial Mixing With Disentangled Representations

Opening the Black Box: Hierarchical Sampling Optimization for Estimating Human Hand Pose

DeepContext: Context-Encoding Neural Pathways for 3D Holistic Scene Understanding

Raster-To-Vector: Revisiting Floorplan Transformation

Realistic Dynamic Facial Textures From a Single Image Using GANs

Scalable Verified Training for Provably Robust Image Classification

Efficient Non-greedy Optimization of Decision Trees

Deep Convolutional Inverse Graphics Network

NeurIPS 2015arXiv

Learning to See Physics via Visual De-animation

Towards Robust Image Classification Using Sequential Attention Models

Computationally Bounded Retrieval

Sparse Projections for High-Dimensional Binary Codes

Picture: A Probabilistic Programming Language for Scene Perception

The Global Patch Collider

Layered Scene Decomposition via the Occlusion-CRF

Neural Scene De-Rendering

Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding

A Unified View of Piecewise Linear Neural Network Verification

Adversarial Robustness through Local Linearization

Learning Transferable Graph Exploration

Are Labels Required for Improving Adversarial Robustness?

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

Training Generative Adversarial Networks by Solving Ordinary Differential Equations

The Autoencoding Variational Autoencoder

Learning Continuous Semantic Representations of Symbolic Expressions

RobustFill: Neural Program Learning under Noisy I/O

Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning

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

Batched High-dimensional Bayesian Optimization via Structural Kernel Learning

Adversarial Risk and the Dangers of Evaluating Against Weak Attacks

Programmatically Interpretable Reinforcement Learning

Structured agents for physical construction

CompILE: Compositional Imitation Learning and Execution

Graph Matching Networks for Learning the Similarity of Graph Structured Objects