Copertina

The good, the bad and the ugly sides of data augmentation: An implicit spectral regularization perspective

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Lin, Chi-Heng, Kaushik, Chiraag, Dyer, Eva L., Muthukumar, Vidya
Natura:	Preprint
Pubblicazione:	2022
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Precise asymptotics of reweighted least-squares algorithms for linear diagonal networks
di: Kaushik, Chiraag, et al.
Pubblicazione: (2024)

A general technique for approximating high-dimensional empirical kernel matrices
di: Kaushik, Chiraag, et al.
Pubblicazione: (2025)

Balanced Data, Imbalanced Spectra: Unveiling Class Disparities with Spectral Imbalance
di: Kaushik, Chiraag, et al.
Pubblicazione: (2024)

Neural topology optimization: the good, the bad, and the ugly
di: Sanu, Suryanarayanan Manoj, et al.
Pubblicazione: (2024)

New Equivalences Between Interpolation and SVMs: Kernels and Structured Features
di: Kaushik, Chiraag, et al.
Pubblicazione: (2023)

Decomposing multimodal embedding spaces with group-sparse autoencoders
di: Kaushik, Chiraag, et al.
Pubblicazione: (2026)

Energy estimates for the good-bad-ugly model
di: Duarte, Miguel
Pubblicazione: (2025)

General Loss Functions Lead to (Approximate) Interpolation in High Dimensions
di: Lai, Kuo-Wei, et al.
Pubblicazione: (2023)

Sharp analysis of out-of-distribution error for "importance-weighted" estimators in the overparameterized regime
di: Lai, Kuo-Wei, et al.
Pubblicazione: (2024)

Improved and Oracle-Efficient Online $\ell_1$-Multicalibration
di: Ghuge, Rohan, et al.
Pubblicazione: (2025)

Task Shift: From Classification to Regression in Overparameterized Linear Models
di: LaBonte, Tyler, et al.
Pubblicazione: (2025)

Estimating stationary mass, frequency by frequency
di: Nakul, Milind, et al.
Pubblicazione: (2025)

Just Wing It: Near-Optimal Estimation of Missing Mass in a Markovian Sequence
di: Pananjady, Ashwin, et al.
Pubblicazione: (2024)

Your contrastive learning problem is secretly a distribution alignment problem
di: Chen, Zihao, et al.
Pubblicazione: (2025)

Editorial April 2025: The protocolisation of evidence synthesis: The good, the bad and the ugly
di: Yasuko Maeda, et al.
Pubblicazione: (2025)

Efficient learning by implicit exploration in bandit problems with side observations
di: Kocak, Tomas, et al.
Pubblicazione: (2026)

On the Unreasonable Effectiveness of Last-layer Retraining
di: Hill, John C., et al.
Pubblicazione: (2025)

Faster Margin Maximization Rates for Generic and Adversarially Robust Optimization Methods
di: Wang, Guanghui, et al.
Pubblicazione: (2023)

Is machine learning good or bad for the natural sciences?
di: Hogg, David W., et al.
Pubblicazione: (2024)

Pre-trained Tabular Foundation Models as Versatile Summary Networks for Neural Posterior Estimation
di: Pickens, Elliot, et al.
Pubblicazione: (2026)

How Does the ReLU Activation Affect the Implicit Bias of Gradient Descent on High-dimensional Neural Network Regression?
di: Lai, Kuo-Wei, et al.
Pubblicazione: (2026)

Connected greedy colourings of perfect graphs and other classes: the good, the bad and the ugly
di: Beaudou, Laurent, et al.
Pubblicazione: (2021)

An asymptotic systems approach for the good-bad-ugly model with application to general relativity
di: Duarte, Miguel, et al.
Pubblicazione: (2025)

The role of TV documentaries to disseminate osteoarchaeology more widely: The good, the bad, and the ugly
di: Piers D. Mitchell
Pubblicazione: (2024)

Causal and Federated Multimodal Learning for Cardiovascular Risk Prediction under Heterogeneous Populations
di: Kaushik, Rohit, et al.
Pubblicazione: (2026)

The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
di: LaBonte, Tyler, et al.
Pubblicazione: (2024)

Last-iterate Convergence for Symmetric, General-sum, $2 \times 2$ Games Under The Exponential Weights Dynamic
di: Wang, Guanghui, et al.
Pubblicazione: (2025)

Deep linear networks for regression are implicitly regularized towards flat minima
di: Marion, Pierre, et al.
Pubblicazione: (2024)

CutisAI: Deep Learning Framework for Automated Dermatology and Cancer Screening
di: Kaushik, Rohit, et al.
Pubblicazione: (2026)

Understanding and inverse design of implicit bias in stochastic learning: a geometric perspective
di: Aladrah, Nicola, et al.
Pubblicazione: (2026)

A Koopman-Bayesian Framework for High-Fidelity, Perceptually Optimized Haptic Surgical Simulation
di: Kaushik, Rohit, et al.
Pubblicazione: (2026)

Power side-channel leakage localization through adversarial training of deep neural networks
di: Gammell, Jimmy, et al.
Pubblicazione: (2024)

Orthogonal projection-based regularization for efficient model augmentation
di: Györök, Bendegúz M., et al.
Pubblicazione: (2025)

Optimal ridge penalty for real-world high-dimensional data can be zero or negative due to the implicit ridge regularization
di: Kobak, Dmitry, et al.
Pubblicazione: (2018)

Position: Message-passing and spectral GNNs are two sides of the same coin
di: Vasileiou, Antonis, et al.
Pubblicazione: (2026)

CNN-LSTM Hybrid Deep Learning Model for Remaining Useful Life Estimation
di: G, Muthukumar, et al.
Pubblicazione: (2024)

The data augmentation algorithm
di: Roy, Vivekananda, et al.
Pubblicazione: (2024)

Revealing Potential Biases in LLM-Based Recommender Systems in the Cold Start Setting
di: Andre, Alexandre, et al.
Pubblicazione: (2025)

The good, the bad and the ugly of pain in haemophilia: Recent evidence on the epidemiology, molecular mechanisms and knowledge gaps preventing optimal treatment
di: Silvia Benemei, et al.
Pubblicazione: (2024)

GraphFM: A generalist graph transformer that learns transferable representations across diverse domains
di: Lachi, Divyansha, et al.
Pubblicazione: (2024)