Native Robustness

Native robustness in machine learning focuses on developing models inherently resistant to various forms of input perturbations, including adversarial attacks and noisy data, without relying solely on post-hoc defenses. Current research emphasizes techniques like ensemble methods, reprogramming existing models, and modifying training procedures (e.g., using different learning rates for specific model layers or incorporating regularization methods) to improve robustness across diverse model architectures, including convolutional neural networks, vision transformers, and large language models. This field is crucial for deploying reliable AI systems in safety-critical applications, such as healthcare and autonomous driving, where model resilience to unexpected inputs is paramount.

Papers

October 29, 2023

Adversarial Examples Are Not Real Features
Ang Li, Yifei Wang, Yiwen Guo, Yisen Wang
Native Robustness Adversarial Example Adversarial Vulnerability Non Robust Feature

October 27, 2023

October 25, 2023

October 23, 2023

Machine Learning and Knowledge: Why Robustness Matters
Jonathan Vandenburgh
Machine Learning Native Robustness Inherent Interpretability Knowledge Based Epistemic Space Model Reliability

October 20, 2023

Assumption violations in causal discovery and the robustness of score matching
Francesco Montagna, Atalanti A. Mastakouri, Elias Eulig, Nicoletta Noceti, Lorenzo Rosasco, Dominik Janzing, Bryon Aragam, Francesco Locatello
Native Robustness Causal Discovery Causal Structure Score Matching Modeling Assumption Causal Discovery Method Observational Causal

October 19, 2023

Robust multimodal models have outlier features and encode more concepts
Jonathan Crabbé, Pau Rodríguez, Vaishaal Shankar, Luca Zappella, Arno Blaas
Native Robustness Large Multimodal Model Representation Space Robust Model Robust Multimodal Outlier Explanation

October 17, 2023

October 16, 2023

October 14, 2023

ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models
Alex Mei, Sharon Levy, William Yang Wang
Native Robustness Red Teaming Robustness Evaluation AI Safety Probabilistic Safety Human Assertion

October 13, 2023

Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning
Qiming Bao, Gael Gendron, Alex Yuxuan Peng, Wanjun Zhong, Neset Tan, Yang Chen, Michael Witbrock, Jiamou Liu
Large Language Model Native Robustness Data Augmentation Logical Reasoning Logical Reasoning Datasets Mixed Task

Native Robustness

Papers

Adversarial Examples Are Not Real Features

Weighted Sampled Split Learning (WSSL): Balancing Privacy, Robustness, and Fairness in Distributed Learning Environments

Robustness of Algorithms for Causal Structure Learning to Hyperparameter Choice

How Robust is Federated Learning to Communication Error? A Comparison Study Between Uplink and Downlink Channels

Back Transcription as a Method for Evaluating Robustness of Natural Language Understanding Models to Speech Recognition Errors

Hierarchical Randomized Smoothing

ELM Ridge Regression Boosting

Fine tuning Pre trained Models for Robustness Under Noisy Labels

Improving Robustness and Reliability in Medical Image Classification with Latent-Guided Diffusion and Nested-Ensembles

Machine Learning and Knowledge: Why Robustness Matters

Assumption violations in causal discovery and the robustness of score matching

Robust multimodal models have outlier features and encode more concepts

Algorithmic Robustness

Non-ergodicity in reinforcement learning: robustness via ergodicity transformations

Understanding Contrastive Learning via Distributionally Robust Optimization

Spoofing Attack Detection in the Physical Layer with Robustness to User Movement

Quantifying Assistive Robustness Via the Natural-Adversarial Frontier

Matching the Neuronal Representations of V1 is Necessary to Improve Robustness in CNNs with V1-like Front-ends

ASSERT: Automated Safety Scenario Red Teaming for Evaluating the Robustness of Large Language Models

Assessing and Enhancing the Robustness of Large Language Models with Task Structure Variations for Logical Reasoning