Search - MIT-IBM Watson AI Lab

Understanding and Improving Visual Prompting: A Label-Mapping Perspective

CVPR

Attacking Visual Language Grounding with Adversarial Examples: A Case Study on Neural Image Captioning

Robustness Natural Language Processing

Adversarial T-shirt! Evading Person Detectors in A Physical World

ECCV Computer Vision

Compressed Decentralized Proximal Stochastic Gradient Method for Nonconvex Composite Problems with Heterogeneous Data

ICML

Topology Attack and Defense for Graph Neural Networks: An Optimization Perspective

IJCAI

Attacking Visual Language Grounding with Adversarial Examples: A Case Study on Neural Image Captioning.

signSGD via Zeroth-Order Oracle

Optimization Robustness

AutoZOOM: Autoencoder-based Zeroth Order Optimization Method for Attacking Black-box Neural Networks

Robustness Optimization

Mod-Squad: Designing Mixtures of Experts As Modular Multi-Task Learners

CVPR

Sign-OPT: Defending the hard-label black-box cyber attack

Cybersecurity Adversarial Machine Learning

Towards Certificated Model Robustness Against Weight Perturbations

AAAI

Fast Training of Provably Robust Neural Networks by SingleProp

AAAI

S3-NeRF: Neural Reflectance Field from Shading and Shadow under a Single Viewpoint

NeurIPS Computer Vision

Structured Adversarial Attack: Towards General Implementation and Better Interpretability

Robustness Explainability

Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approach

Robustness

Zeroth-Order Stochastic Variance Reduction for Nonconvex Optimization

Optimization Deep Learning

Neural Network Robustness Certification with General Activation Functions

Robustness Deep Learning

Joint Edge-Model Sparse Learning is Provably Efficient for Graph Neural Networks

ICLR

A Theoretical Understanding of Shallow Vision Transformers: Learning, Generalization, and Sample Complexity

ICLR

Efficient Neural Network Robustness Certification with General Activation Functions

NeurIPS

Fast Incremental von Neumann Graph Entropy Computation: Theory, Algorithm, and Applications

ICML

Proper Network Interpretability Helps Adversarial Robustness in Classification

ICML

How unlabeled data improve generalization in self-training? A one-hidden-layer theoretical analysis

ICLR

CNN-Cert: An Efficient Framework for Certifying Robustness of Convolutional Neural Networks

Robustness

Practical Detection of Trojan Neural Networks: Data-Limited and Data-Free Cases

ECCV

On the Design of Black-box Adversarial Examples by Leveraging Gradient-free Optimization and Operator Splitting Method

Optimization Robustness

Higher-Order Certification For Randomized Smoothing

Optimization

Hidden Cost of Randomized Smoothing

AISTATS Machine Learning

Self-Progressing Robust Training

AAAI Machine Learning

PROVEN: Verifying Robustness of Neural Networks with a Probabilistic Approach

ICML

On Fast Adversarial Robustness Adaptation in Model-Agnostic Meta-Learning

ICLR Machine Learning

Fast Learning of Graph Neural Networks with Guaranteed Generalizability: One-hidden-layer Case

ICML Machine Learning

Is There a Trade-Off Between Fairness and Accuracy? A Perspective Using Mismatched Hypothesis Testing

ICML Machine Learning

Adversarial Attack Generation Empowered by Min-Max Optimization

NeurIPS

When does Contrastive Learning Preserve Adversarial Robustness from Pretraining to Finetuning?

NeurIPS

Why Lottery Ticket Wins? A Theoretical Perspective of Sample Complexity on Sparse Neural Networks

NeurIPS

A Gromov-Wasserstein Geometric View of Spectrum-Preserving Graph Coarsening

ICML

SparseViT: Revisiting Activation Sparsity for Efficient High-Resolution Vision Transformer

CVPR

On-Device Training Under 256KB Memory

NeurIPS

Masked Motion Encoding for Self-Supervised Video Representation Learning

CVPR

Visual Dependency Transformers: Dependency Tree Emerges from Reversed Attention

CVPR

Topological Experience Replay

ICLR

Multimodal Clustering Networks for Self-Supervised Learning From Unlabeled Videos

ICCV

CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification

ICCV

Adversarial Robustness: From Self-Supervised Pre-Training to Fine-Tuning

CVPR

Deep Analysis of CNN-based Spatio-temporal Representations for Action Recognition

CVPR Computer Vision

The Lottery Tickets Hypothesis for Supervised and Self-supervised Pre-training in Computer Vision Models

CVPR Computer Vision

Location-aware Graph Convolutional Networks for Video Question Answering

AAAI

Experiences and Insights for Collaborative Industry-Academic Research in Artificial Intelligence

Entrepreneurship Future of Work

A Sequential Set Generation Method for Predicting Set-Valued Outputs

AAAI

Training Stronger Baselines for Learning to Optimize

Optimization

MCUNet: Tiny Deep Learning on IoT Devices

Deep Learning Efficient AI

Towards Verifying Robustness of Neural Networks against Semantic Perturbations

CVPR

Revisiting Contrastive Learning through the Lens of Neighborhood Component Analysis: an Integrated Framework

ICML

Rate-improved inexact augmented Lagrangian method for constrained nonconvex optimization

AISTATS

Detector-Free Weakly Supervised Grounding by Separation

ICCV

Data-Efficient Double-Win Lottery Tickets from Robust Pre-training

ICML

Procedural Image Programs for Representation Learning

NeurIPS

SNAKE: Shape-aware Neural 3D Keypoint Field

NeurIPS

A Broad Study on the Transferability of Visual Representations With Contrastive Learning

ICCV

Learning Active Camera for Multi-Object Navigation

NeurIPS

Dense Regression Network for Video Grounding

CVPR

VALHALLA: Visual Hallucination for Machine Translation

CVPR

Task2Sim: Towards Effective Pre-training and Transfer from Synthetic Data

CVPR

LitePose: Efficient Architecture Design for 2D Human Pose Estimation

CVPR

Weakly-Supervised Multi-Granularity Map Learning for Vision-and-Language Navigation

NeurIPS

Optimizer Amalgamation

ICLR

ComPhy: Compositional Physical Reasoning of Objects and Events from Videos

ICLR

Planning with Large Language Models for Code Generation

ICLR

Discrete Graph Structure Learning for Forecasting Multiple Time Series

ICLR Time Series

Min-Max Optimization without Gradients: Convergence and Applications to Black-Box Evasion and Poisoning Attacks

ICML

DAG-GNN: DAG Structure Learning with Graph Neural Networks

ICML

ZO-AdaMM: Zeroth-Order Adaptive Momentum Method for Black-Box Optimization

NeurIPS

CAG: A Real-Time Low-Cost Enhanced-Robustness High-Transferability Content-Aware Adversarial Attack Generator

AAAI

Proximal Stochastic Recursive Momentum Methods for Nonconvex Composite Decentralized Optimization

AAAI

Redeeming Intrinsic Rewards via Constrained Optimization

NeurIPS Optimization

DexDeform: Dexterous Deformable Object Manipulation with Human Demonstrations and Differentiable Physics

ICLR

PAC-NeRF: Physics Augmented Continuum Neural Radiance Fields for Geometry-Agnostic System Identification

ICLR

Graph Neural Network-Inspired Kernels for Gaussian Processes in Semi-Supervised Learning

ICLR

3D Concept Learning and Reasoning from Multi-View Images

CVPR

Pic2Word: Mapping Pictures to Words for Zero-shot Composed Image Retrieval

CVPR

GC-Flow: A Graph-Based Flow Network for Effective Clustering

ICML

Hierarchical Grammar-Induced Geometry for Data-Efficient Molecular Property Prediction

ICML

Advancing Model Pruning via Bi-level Optimization

NeurIPS Optimization

EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphs

Graph Deep Learning AAAI

Exponentially Improving the Complexity of Simulating the Weisfeiler-Lehman Test with Graph Neural Networks

NeurIPS

RT3D: Achieving Real-Time Execution of 3D Convolutional Neural Networks on Mobile Devices

AAAI Machine Learning

Constrained Generation of Semantically Valid Graphs via Regularizing Variational Autoencoders

Generative Models Graph Deep Learning

Scalable Graph Learning for Anti-Money Laundering: A First Look

AI in Finance Graph Deep Learning

On the Convergence of A Class of Adam-Type Algorithms for Non-Convex Optimization

Optimization

Anti-Money Laundering in Bitcoin: Experimenting with Graph Convolutional Networks for Financial Forensics

AI in Finance Graph Deep Learning

Self-supervised Moving Vehicle Tracking with Stereo Sound

Multimodal Learning Computer Vision

Big-Little-Video-Net: Work smarter, not harder, for video understanding

Computer Vision

ZO-AdaMM: Derivative-free optimization for black-box problems

Optimization NeurIPS

Deep Symbolic Superoptimization Without Human Knowledge

Neuro-Symbolic AI Optimization

The Lottery Ticket Hypothesis for the Pre-trained BERT Networks

Efficient AI

AdaMML: Adaptive Multi-Modal Learning for Efficient Video Recognition

ICCV Multimodal Learning

RSPNet: Relative Speed Perception for Unsupervised Video Representation Learning.

AAAI Computer Vision

Grounding Physical Concepts of Objects and Events Through Dynamic Visual Reasoning

ICLR Computer Vision

Robust Overfitting may be mitigated by properly learned smoothening

ICLR

Debiased Contrastive Learning

NeurIPS

Foley Music: Learning to Generate Music from Videos

ECCV Computer Vision

CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks

NeurIPS

Dynamic Video Quantization for Efficient Inference

ICCV ICCV

NASTransfer: Analyzing Architecture Transferability in Large Scale Neural Architecture Search

AAAI Computer Vision

STAR: A Benchmark for Situated Reasoning in Real-World Videos

NeurIPS

Dynamic Visual Reasoning by Learning Differentiable Physics Models from Video and Language

NeurIPS

Memory-efficient Patch-based Inference for Tiny Deep Learning

NeurIPS

Everything at Once – Multi-modal Fusion Transformer for Video Retrieval

CVPR

On the Equivalence between Neural Network and Support Vector Machine

NeurIPS

Dynamic Distillation Network for Cross-Domain Few-Shot Recognition with Unlabeled Data

NeurIPS

Graph Universal Adversarial Attacks: A Few Bad Actors Ruin Graph Learning Models

Machine Learning Social Networks

Unsupervised Learning of Graph Hierarchical Abstractions with Differentiable Coarsening and Optimal Transport

AAAI Machine Learning

An Image Enhancing Pattern-based Sparsity for Real-time Inference on Mobile Devices

ECCV Computer Vision

Long Live the Lottery: The Existence of Winning Tickets in Lifelong Learning

ICLR

A content-aware attack generator for AI cybersecurity

Adversarial Machine Learning Cybersecurity

Embedding Compression with Isotropic Iterative Quantization

Natural Language Processing

Online AI planning with graph neural networks and adaptive scheduling

Automated Planning Graph Deep Learning

Learning to learn with distributional signatures for text data

ICLR Transfer Learning

Fast and efficient black-box testing for AI cybersecurity

Cybersecurity Robustness

Graph-Augmented Normalizing Flows for Anomaly Detection of Multiple Time Series

ICLR

RegionViT: Regional-to-Local Attention for Vision Transformers

ICLR

Can an Image Classifier Suffice For Action Recognition?

ICLR

Data-Efficient Graph Grammar Learning for Molecular Generation

ICLR

Directed Acyclic Graph Neural Networks

ICLR Graph Deep Learning