Matches in SemOpenAlex for { <https://semopenalex.org/work/W4308243025> ?p ?o ?g. }
Showing items 1 to 54 of
54
with 100 items per page.
- W4308243025 abstract "The high cost of acquiring and annotating samples has made the `few-shot' learning problem of prime importance. Existing works mainly focus on improving performance on clean data and overlook robustness concerns on the data perturbed with adversarial noise. Recently, a few efforts have been made to combine the few-shot problem with the robustness objective using sophisticated Meta-Learning techniques. These methods rely on the generation of adversarial samples in every episode of training, which further adds a computational burden. To avoid such time-consuming and complicated procedures, we propose a simple but effective alternative that does not require any adversarial samples. Inspired by the cognitive decision-making process in humans, we enforce high-level feature matching between the base class data and their corresponding low-frequency samples in the pretraining stage via self distillation. The model is then fine-tuned on the samples of novel classes where we additionally improve the discriminability of low-frequency query set features via cosine similarity. On a 1-shot setting of the CIFAR-FS dataset, our method yields a massive improvement of $60.55%$ & $62.05%$ in adversarial accuracy on the PGD and state-of-the-art Auto Attack, respectively, with a minor drop in clean accuracy compared to the baseline. Moreover, our method only takes $1.69times$ of the standard training time while being $approx$ $5times$ faster than state-of-the-art adversarial meta-learning methods. The code is available at https://github.com/vcl-iisc/robust-few-shot-learning." @default.
- W4308243025 created "2022-11-09" @default.
- W4308243025 creator A5030601126 @default.
- W4308243025 creator A5031349314 @default.
- W4308243025 creator A5037927566 @default.
- W4308243025 creator A5091123666 @default.
- W4308243025 date "2022-11-03" @default.
- W4308243025 modified "2023-09-27" @default.
- W4308243025 title "Robust Few-shot Learning Without Using any Adversarial Samples" @default.
- W4308243025 doi "https://doi.org/10.48550/arxiv.2211.01598" @default.
- W4308243025 hasPublicationYear "2022" @default.
- W4308243025 type Work @default.
- W4308243025 citedByCount "0" @default.
- W4308243025 crossrefType "posted-content" @default.
- W4308243025 hasAuthorship W4308243025A5030601126 @default.
- W4308243025 hasAuthorship W4308243025A5031349314 @default.
- W4308243025 hasAuthorship W4308243025A5037927566 @default.
- W4308243025 hasAuthorship W4308243025A5091123666 @default.
- W4308243025 hasBestOaLocation W43082430251 @default.
- W4308243025 hasConcept C104317684 @default.
- W4308243025 hasConcept C119857082 @default.
- W4308243025 hasConcept C153180895 @default.
- W4308243025 hasConcept C154945302 @default.
- W4308243025 hasConcept C185592680 @default.
- W4308243025 hasConcept C37736160 @default.
- W4308243025 hasConcept C41008148 @default.
- W4308243025 hasConcept C55493867 @default.
- W4308243025 hasConcept C63479239 @default.
- W4308243025 hasConceptScore W4308243025C104317684 @default.
- W4308243025 hasConceptScore W4308243025C119857082 @default.
- W4308243025 hasConceptScore W4308243025C153180895 @default.
- W4308243025 hasConceptScore W4308243025C154945302 @default.
- W4308243025 hasConceptScore W4308243025C185592680 @default.
- W4308243025 hasConceptScore W4308243025C37736160 @default.
- W4308243025 hasConceptScore W4308243025C41008148 @default.
- W4308243025 hasConceptScore W4308243025C55493867 @default.
- W4308243025 hasConceptScore W4308243025C63479239 @default.
- W4308243025 hasLocation W43082430251 @default.
- W4308243025 hasLocation W43082430252 @default.
- W4308243025 hasOpenAccess W4308243025 @default.
- W4308243025 hasPrimaryLocation W43082430251 @default.
- W4308243025 hasRelatedWork W2952919291 @default.
- W4308243025 hasRelatedWork W3035729345 @default.
- W4308243025 hasRelatedWork W3114774860 @default.
- W4308243025 hasRelatedWork W3208304128 @default.
- W4308243025 hasRelatedWork W4287549955 @default.
- W4308243025 hasRelatedWork W4306177400 @default.
- W4308243025 hasRelatedWork W4311734044 @default.
- W4308243025 hasRelatedWork W4379255972 @default.
- W4308243025 hasRelatedWork W4379258830 @default.
- W4308243025 hasRelatedWork W4383955378 @default.
- W4308243025 isParatext "false" @default.
- W4308243025 isRetracted "false" @default.
- W4308243025 workType "article" @default.