A farewell to the bias-variance tradeoff? an overview of the theory of overparameterized machine learning
Y Dar, V Muthukumar, RG Baraniuk - arxiv preprint arxiv:2109.02355, 2021 - arxiv.org
The rapid recent progress in machine learning (ML) has raised a number of scientific
questions that challenge the longstanding dogma of the field. One of the most important …
questions that challenge the longstanding dogma of the field. One of the most important …
High-dimensional analysis of double descent for linear regression with random projections
We consider linear regression problems with a varying number of random projections,
where we provably exhibit a double descent curve for a fixed prediction problem, with a high …
where we provably exhibit a double descent curve for a fixed prediction problem, with a high …
Probing transfer learning with a model of synthetic correlated datasets
Transfer learning can significantly improve the sample efficiency of neural networks, by
exploiting the relatedness between a data-scarce target task and a data-abundant source …
exploiting the relatedness between a data-scarce target task and a data-abundant source …
Universality in Transfer Learning for Linear Models
We study the problem of transfer learning and fine-tuning in linear models for both
regression and binary classification. In particular, we consider the use of stochastic gradient …
regression and binary classification. In particular, we consider the use of stochastic gradient …
Fast ocean front detection using deep learning edge detection models
Small-scale ocean fronts play a significant role in absorbing the excess heat and CO 2
generated by climate change, yet their dynamics are not well understood. The existing in situ …
generated by climate change, yet their dynamics are not well understood. The existing in situ …
Improved linear regression prediction by transfer learning
Transfer learning, also referred as knowledge transfer, aims at reusing knowledge from a
source dataset to a similar target one. While several studies address the problem of what to …
source dataset to a similar target one. While several studies address the problem of what to …
Frozen Overparameterization: A Double Descent Perspective on Transfer Learning of Deep Neural Networks
We study the generalization behavior of transfer learning of deep neural networks (DNNs).
We adopt the overparameterization perspective--featuring interpolation of the training data …
We adopt the overparameterization perspective--featuring interpolation of the training data …
Seasonal variability of the Pearl River Plume front based on deep learning
Z Chen, J Feng, T Li, S Zhang, Q Lian… - Continental Shelf Research, 2025 - Elsevier
Ocean fronts are characterized by active vertical shear and mixing, which play a role in
primary production, fishery resources, and marine ecosystems. While oceanographers have …
primary production, fishery resources, and marine ecosystems. While oceanographers have …
TL-PCA: Transfer Learning of Principal Component Analysis
Principal component analysis (PCA) can be significantly limited when there is too few
examples of the target data of interest. We propose a transfer learning approach to PCA (TL …
examples of the target data of interest. We propose a transfer learning approach to PCA (TL …
Impact of regularization on the effectiveness of transfer learning for olive disease classification
Transfer learning is a machine learning method where a model developed for a task is
reused as the starting point for a model on a second task. This work presents an approach …
reused as the starting point for a model on a second task. This work presents an approach …