26 citations · 42 across the 20 of their papers we have counts for
9 papers · 1 filter
Online Feature Updates Improve Online (Generalized) Label Shift Adaptation
Ruihan Wu, Siddhartha Datta, Yi Su +3
This paper addresses the prevalent issue of label shift in an online setting with missing labels, where data distributions change over time and obtaining timely labels is challengi…
Projected Subnetworks Scale Adaptation
Siddhartha Datta, Nigel Shadbolt
Large models support great zero-shot and few-shot capabilities. However, updating these models on new tasks can break performance on previous seen tasks and their zero/few-shot uns…
Multiple Modes for Continual Learning
Siddhartha Datta, Nigel Shadbolt
Adapting model parameters to incoming streams of data is a crucial factor to deep learning scalability. Interestingly, prior continual learning strategies in online settings inadve…
Interpolating Compressed Parameter Subspaces
Siddhartha Datta, Nigel Shadbolt
Inspired by recent work on neural subspaces and mode connectivity, we revisit parameter subspace sampling for shifted and/or interpolatable input distributions (instead of a single…
Learn2Weight: Parameter Adaptation against Similar-domain Adversarial Attacks
Siddhartha Datta
Recent work in black-box adversarial attacks for NLP systems has attracted much attention. Prior black-box attacks assume that attackers can observe output labels from target model…
Low-Loss Subspace Compression for Clean Gains against Multi-Agent Backdoor Attacks
Siddhartha Datta, Nigel Shadbolt
Recent exploration of the multi-agent backdoor attack demonstrated the backfiring effect, a natural defense against backdoor attacks where backdoored inputs are randomly classified…
Backdoors Stuck At The Frontdoor: Multi-Agent Backdoor Attacks That Backfire
Siddhartha Datta, Nigel Shadbolt
Malicious agents in collaborative learning and outsourced data collection threaten the training of clean models. Backdoor attacks, where an attacker poisons a model during training…
Hiding Behind Backdoors: Self-Obfuscation Against Generative Models
Siddhartha Datta, Nigel Shadbolt
Attack vectors that compromise machine learning pipelines in the physical world have been demonstrated in recent research, from perturbations to architectural components. Building…
Widen The Backdoor To Let More Attackers In
Siddhartha Datta, Giulio Lovisotto, Ivan Martinovic +1
As collaborative learning and the outsourcing of data collection become more common, malicious actors (or agents) which attempt to manipulate the learning process face an additiona…