Meta-learning approaches for few-shot learning: A survey of recent advances

H Gharoun, F Momenifar, F Chen… - ACM Computing …, 2024 - dl.acm.org
Despite its astounding success in learning deeper multi-dimensional data, the performance
of deep learning declines on new unseen tasks mainly due to its focus on same-distribution …

Few-shot open-set learning for on-device customization of keyword spotting systems

M Rusci, T Tuytelaars - ar** for few-shot fine-grained visual classification
Q Wu, T Song, S Fan, Z Chen, K **, H Zhou - Image and Vision Computing, 2024 - Elsevier
Few-shot fine-grained visual classification aims to identify fine-grained concepts with very
few samples, which is widely used in many fields, such as the classification of different …

Task-agnostic open-set prototype for few-shot open-set recognition

B Kim, JT Lee, K Shim, S Chang - 2023 IEEE International …, 2023 - ieeexplore.ieee.org
In few-shot open-set recognition (FSOSR), a network learns to recognize closed-set samples
with a few support samples while rejecting open-set samples with no class cue. Unlike …

Joint embedding learning and latent subspace probing for cross-domain few-shot keyword spotting

M Ozay - ICASSP 2024-2024 IEEE International Conference on …, 2024 - ieeexplore.ieee.org
Probing classifiers (PCs) have been employed as one of the notable approaches for
exploring properties of deep neural network (DNN) models in various tasks such as natural …

Self-Learning for Personalized Keyword Spotting on Ultra-Low-Power Audio Sensors

M Rusci, F Paci, M Fariselli, E Flamand… - IEEE Internet of …, 2024 - ieeexplore.ieee.org
This paper proposes a self-learning method to incrementally train (fine-tune) a personalized
Keyword Spotting (KWS) model after the deployment on ultra-low power smart audio …

Improving small footprint few-shot keyword spotting with supervision on auxiliary data

S Yang, B Kim, K Shim, S Chang - arxiv preprint arxiv:2309.00647, 2023 - arxiv.org
Few-shot keyword spotting (FS-KWS) models usually require large-scale annotated datasets
to generalize to unseen target keywords. However, existing KWS datasets are limited in …

Fully unsupervised training of few-shot keyword spotting

D Lee, M Kim, SH Mun, MH Han… - 2022 IEEE Spoken …, 2023 - ieeexplore.ieee.org
For training a few-shot keyword spotting (FS-KWS) model, a large labeled dataset
containing massive target keywords has known to be essential to generalize to arbitrary …

Unlocking Transfer Learning for Open-World Few-Shot Recognition

B Kim, J Lee, K Shim, S Chang - arxiv preprint arxiv:2411.09986, 2024 - arxiv.org
Few-Shot Open-Set Recognition (FSOSR) targets a critical real-world challenge, aiming to
categorize inputs into known categories, termed closed-set classes, while identifying open …