refafactory.blogg.se

( paper) MetaGAN An Adversarial Approach to Few-Shot Learning.One line descriptions: Generate features to expasion small datasets to large datasets, then fintune. Teach to use tradional machining learning methods.Meta-learning with differentiable closed-form solvers Delta-encoder: an effective sample synthesis method for few-shot object recognition ( paper code) Generating Classification Weights with GNN Denoising Autoencoders for Few-Shot Learning High dimensional problem is hard to solve in the low-data circumstances, so this work try to bypass the limitations by learning a data-dependent latent low-dimensional latent space of model parameters.(RECOMMENDED!) ( paper code) Meta-learning with latent embedding optimization Using normal way to pretrain the backbone on the base classes, then using the base class weights to fintune the classifier on the few-shot episodic network.( paper code) Incremental Few-Shot Learning with Attention Attractor Networks Use a meta-learner to generate parameters for the feature extractor.( paper code) TAFE-Net- Task-Aware Feature Embeddings for Low Shot Learning ( paper code) LGM-Net: Learning to Generate Matching Networks for Few shot Learning ( paper) Dense Classification and Implanting for Few-Shot Learning (RECOMMENDED!) ( paper) Learning to Learn Image Classifiers with Visual Analogy

(I DONOT like it, the exp only compare to matching networks and "Generative + classifier") Passing, generate weights for classifier.( paper) Low-Shot Learning with Imprinted Weights ( paper code) Dynamic Few-Shot Visual Learning without Forgetting Improve the Meta-Learner LSTM, by adding temporal convolution and caual attention to the network.( paper) A Simple Neural Attentive Meta- Learner ( paper) On First-Order Meta-Learning Algorithms Use LSTM to generate classifier's parameters.( paper code) Optimization as a Model for Few-shot Learning One line descriptions: Generate parameters for the classifier or finetune part of the models Papers ( paper code) A Closer Look At Few-shot Classification.( paper) Generalizing from a Few Examples A Survey on Few-Shot Learning.