Nicholas Donald Lane

Chief Scientific Officer@Flower Labs·英国·OpenReview

研究方向

Federated Learning · Neural Accelerator Design and Tooling · Deep Learning · Model Compression · Machine Learning Systems · Mobile Computing · Neural Architecture Search · Activity Recognition

8.0

CLUES: Collaborative Private-domain High-quality Data Selection for LLMs via Training Dynamics

NeurIPS 2024Poster

通讯

4.8

Towards Neural Architecture Search through Hierarchical Generative Modeling

ICLR 2024Rejected

4.3

Breaking Physical and Linguistic Borders: Multilingual Federated Prompt Tuning for Low-Resource Languages

ICLR 2024Poster

通讯

3.7

vFedSec: Efficient Secure Aggregation for Vertical Federated Learning via Secure Layer

ICLR 2024Withdrawn

通讯

3.5

FedAnchor: Enhancing Federated Semi-Supervised Learning with Label Contrastive Loss

合作者 (20)

Pedro Porto Buarque de Gusmao

2 篇

William F. Shen

2 篇

Nicholas Donald Lane

DEPT: Decoupled Embeddings for Pre-training Language Models

SparsyFed: Sparse Adaptive Federated Learning

Worldwide Federated Training of Language Models

TinyTrain: Deep Neural Network Training at the Extreme Edge

CLUES: Collaborative Private-domain High-quality Data Selection for LLMs via Training Dynamics

Towards Neural Architecture Search through Hierarchical Generative Modeling

Breaking Physical and Linguistic Borders: Multilingual Federated Prompt Tuning for Low-Resource Languages

vFedSec: Efficient Secure Aggregation for Vertical Federated Learning via Secure Layer

FedAnchor: Enhancing Federated Semi-Supervised Learning with Label Contrastive Loss