Low-Shot Learning from Imaginary Data

01/16/2018
by   Yu-Xiong Wang, et al.
0

Humans can quickly learn new visual concepts, perhaps because they can easily visualize or imagine what novel objects look like from different views. Incorporating this ability to hallucinate novel instances of new concepts might help machine vision systems perform better low-shot learning, i.e., learning concepts from few examples. We present a novel approach to low-shot learning that uses this idea. Our approach builds on recent progress in meta-learning ("learning to learn") by combining a meta-learner with a "hallucinator" that produces additional training examples, and optimizing both models jointly. Our hallucinator can be incorporated into a variety of meta-learners and provides significant gains: up to a 6 point boost in classification accuracy when only a single training example is available, yielding state-of-the-art performance on the challenging ImageNet low-shot classification benchmark.

READ FULL TEXT
research
05/28/2019

Image Deformation Meta-Networks for One-Shot Learning

Humans can robustly learn novel visual concepts even when images undergo...
research
06/09/2016

Low-shot Visual Recognition by Shrinking and Hallucinating Features

Low-shot visual learning---the ability to recognize novel object categor...
research
01/29/2019

Harnessing GANs for Addition of New Classes in VSR

It is an easy task for humans to learn and generalize a problem, perhaps...
research
08/17/2022

Constrained Few-Shot Learning: Human-Like Low Sample Complexity Learning and Non-Episodic Text Classification

Few-shot learning (FSL) is an emergent paradigm of learning that attempt...
research
07/05/2020

MetaConcept: Learn to Abstract via Concept Graph for Weakly-Supervised Few-Shot Learning

Meta-learning has been proved to be an effective framework to address fe...
research
06/17/2020

Improving Few-Shot Visual Classification with Unlabelled Examples

We propose a transductive meta-learning method that uses unlabelled inst...
research
05/02/2023

Accelerating Neural Self-Improvement via Bootstrapping

Few-shot learning with sequence-processing neural networks (NNs) has rec...

Please sign up or login with your details

Forgot password? Click here to reset