paper-with-me

홈 › Papers

Tangent Transformers for Composition, Privacy and Removal

2023-07-16 · Tian Yu Liu, Aditya Golatkar, Stefano Soatto

We introduce Tangent Attention Fine-Tuning (TAFT), a method for fine-tuning linearized transformers obtained by computing a First-order Taylor Expansion around a pre-trained initialization. We show that the Jacobian-Vector Product resulting from linearization can be computed efficiently in a single forward pass, reducing training and inference cost to the same order of magnitude as its original non-linear counterpart, while using the same number of parameters. Furthermore, we show that, when applied to various downstream visual classification tasks, the resulting Tangent Transformer fine-tuned with TAFT can perform comparably with fine-tuning the original non-linear network. Since Tangent Transformers are linear with respect to the new set of weights, and the resulting fine-tuning loss is convex, we show that TAFT enjoys several advantages compared to non-linear fine-tuning when it comes to model composition, parallel training, machine unlearning, and differential privacy. Our code is available at: https://github.com/tianyu139/tangent-model-composition

📄 PDF Abstract BibTeX arXiv:2307.08122

Code (1)

tianyu139/tangent-model-composition 공식 구현 pytorch

Tasks

Machine Unlearning

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Residual Connection 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

Towards Robust Knowledge Removal in Federated Learning with High Data Heterogeneity

2025-10-15 · Riccardo Santi, Riccardo Salami, Simone Calderara arxiv

Nowdays, there are an abundance of portable devices capable of collecting large amounts of data and with decent computational power. This opened the possibility to train AI models in a distributed manner, preserving the …

Federated Learning

Tangent differential privacy

2024-06-06 · Lexing Ying

Differential privacy is a framework for protecting the identity of individual data points in the decision-making process. In this note, we propose a new form of differential privacy called tangent differential privacy. C…

Decision Making

Learning in Infinitesimal Non-Compositional Sketches

2026-07-16 · Sridhar Mahadevan arxiv

This paper develops a categorical framework -- Learning in Infinitesimal Non-Compositional Sketches (LINCS) -- as the repair of non-compositionality: failures of diagrams to factor through quotient sketches lifted to the…

Reinforcement Learning

Privacy at Scale: Introducing the PrivaSeer Corpus of Web Privacy Policies

2020-04-23 · ACL 2021 5 · Mukund Srinath, Shomir Wilson, C. Lee Giles

Organisations disclose their privacy practices by posting privacy policies on their website. Even though users often care about their digital privacy, they often don't read privacy policies since they require a significa…

Document ClassificationKeyphrase Extraction

DiffDecompose: Layer-Wise Decomposition of Alpha-Composited Images via Diffusion Transformers

2025-05-24 · Zitong Wang, Hang Zhao, Qianyu Zhou, Xuequan Lu 외

Diffusion models have recently motivated great success in many generation tasks like object removal. Nevertheless, existing image decomposition methods struggle to disentangle semi-transparent or transparent layer occlus…

Flare Removal