Search Results for author: Siddhartha Datta

Found 16 papers, 3 papers with code

Online Feature Updates Improve Online (Generalized) Label Shift Adaptation

no code implementations5 Feb 2024 Ruihan Wu, Siddhartha Datta, Yi Su, Dheeraj Baby, Yu-Xiang Wang, Kilian Q. Weinberger

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 challenging.

Missing Labels Self-Supervised Learning

Projected Subnetworks Scale Adaptation

no code implementations27 Jan 2023 Siddhartha Datta, Nigel Shadbolt

Large models support great zero-shot and few-shot capabilities.

Cross-Reality Re-Rendering: Manipulating between Digital and Physical Realities

no code implementations15 Nov 2022 Siddhartha Datta

As digital realities become an increasingly-impactful aspect of human lives, we investigate the design of a system that enables users to manipulate the perception of both their physical realities and digital realities.

Multiple Modes for Continual Learning

no code implementations29 Sep 2022 Siddhartha Datta, Nigel Shadbolt

Adapting model parameters to incoming streams of data is a crucial factor to deep learning scalability.

Continual Learning

Interpolating Compressed Parameter Subspaces

no code implementations19 May 2022 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, unshifted distribution).

Continual Learning

Low-Loss Subspace Compression for Clean Gains against Multi-Agent Backdoor Attacks

no code implementations7 Mar 2022 Siddhartha Datta, Nigel Shadbolt

clean labels, which motivates this paper's work on the construction of multi-agent backdoor defenses that maximize accuracy w. r. t.

Backdoor Attack

Backdoors Stuck At The Frontdoor: Multi-Agent Backdoor Attacks That Backfire

no code implementations28 Jan 2022 Siddhartha Datta, Nigel Shadbolt

Malicious agents in collaborative learning and outsourced data collection threaten the training of clean models.

Backdoor Attack backdoor defense

Hiding Behind Backdoors: Self-Obfuscation Against Generative Models

no code implementations24 Jan 2022 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.

BIG-bench Machine Learning

Widen The Backdoor To Let More Attackers In

no code implementations9 Oct 2021 Siddhartha Datta, Giulio Lovisotto, Ivan Martinovic, Nigel Shadbolt

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 additional obstacle as they compete with each other.

Backdoor Attack

I Want My App That Way: Reclaiming Sovereignty Over Personal Devices

1 code implementation23 Feb 2021 Konrad Kollnig, Siddhartha Datta, Max Van Kleek

Dark patterns in mobile apps take advantage of cognitive biases of end-users and can have detrimental effects on people's lives.

Human-Computer Interaction

DeepObfusCode: Source Code Obfuscation Through Sequence-to-Sequence Networks

1 code implementation3 Sep 2019 Siddhartha Datta

The paper explores a novel methodology in source code obfuscation through the application of text-based recurrent neural network (RNN) encoder-decoder models in ciphertext generation and key generation.

Cannot find the paper you are looking for? You can Submit a new open access paper.