Overview frequency principle/spectral bias in deep learning

ZQJ Xu, Y Zhang, T Luo - Communications on Applied Mathematics and …, 2024 - Springer
Understanding deep learning is increasingly emergent as it penetrates more and more into
industry and science. In recent years, a research line from Fourier analysis sheds light on …

Robust heterogeneous federated learning under data corruption

X Fang, M Ye, X Yang - Proceedings of the IEEE/CVF …, 2023 - openaccess.thecvf.com
Abstract Model heterogeneous federated learning is a realistic and challenging problem.
However, due to the limitations of data collection, storage, and transmission conditions, as …

Single chip photonic deep neural network with accelerated training

S Bandyopadhyay, A Sludds, S Krastanov… - arxiv preprint arxiv …, 2022 - arxiv.org
As deep neural networks (DNNs) revolutionize machine learning, energy consumption and
throughput are emerging as fundamental limitations of CMOS electronics. This has …

Word order does matter and shuffled language models know it

M Abdou, V Ravishankar, A Kulmizev… - Proceedings of the 60th …, 2022 - aclanthology.org
Recent studies have shown that language models pretrained and/or fine-tuned on randomly
permuted sentences exhibit competitive performance on GLUE, putting into question the …

Fedfa: Federated feature augmentation

T Zhou, E Konukoglu - arxiv preprint arxiv:2301.12995, 2023 - arxiv.org
Federated learning is a distributed paradigm that allows multiple parties to collaboratively
train deep models without exchanging the raw data. However, the data distribution among …

Noisy recurrent neural networks

SH Lim, NB Erichson, L Hodgkinson… - Advances in Neural …, 2021 - proceedings.neurips.cc
We provide a general framework for studying recurrent neural networks (RNNs) trained by
injecting noise into hidden states. Specifically, we consider RNNs that can be viewed as …

Explicit regularization in overparametrized models via noise injection

A Orvieto, A Raj, H Kersting… - … Conference on Artificial …, 2023 - proceedings.mlr.press
Injecting noise within gradient descent has several desirable features, such as smoothing
and regularizing properties. In this paper, we investigate the effects of injecting noise before …

Noisy feature mixup

SH Lim, NB Erichson, F Utrera, W Xu… - arxiv preprint arxiv …, 2021 - arxiv.org
We introduce Noisy Feature Mixup (NFM), an inexpensive yet effective method for data
augmentation that combines the best of interpolation based training and noise injection …

Neural decoding reveals specialized kinematic tuning after an abrupt cortical transition

RM Glanz, G Sokoloff, MS Blumberg - Cell reports, 2023 - cell.com
The primary motor cortex (M1) exhibits a protracted period of development, including the
development of a sensory representation long before motor outflow emerges. In rats, this …

On the generalization of models trained with SGD: Information-theoretic bounds and implications

Z Wang, Y Mao - arxiv preprint arxiv:2110.03128, 2021 - arxiv.org
This paper follows up on a recent work of Neu et al.(2021) and presents some new
information-theoretic upper bounds for the generalization error of machine learning models …