Deep learning

Pattern recognition (psychology)

Invariant (physics)

Cosine similarity

Mathematics

Mathematical physics

Image (mathematics)

Similarity (geometry)

Competitive learning

Machine learning

Artificial intelligence

Softmax function

Computer science

Unsupervised learning

Embedding

Feature learning

Feature vector

This paper studies the unsupervised embedding learning problem, which requires an effective similarity measurement between samples in low-dimensional embedding space. Motivated by the positive concentrated and negative separated properties observed from category-wise supervised learning, we propose to utilize the instance-wise supervision to approximate these properties, which aims at learning data augmentation invariant and instance spread-out features. To achieve this goal, we propose a novel instance based softmax embedding method, which directly optimizes the `real' instance features on top of the softmax function. It achieves significantly faster learning speed and higher accuracy than all existing methods. The proposed method performs well for both seen and unseen testing categories with cosine similarity. It also achieves competitive performance even without pre-trained network over samples from fine-grained categories.

Unsupervised Embedding Learning via Invariant and Spreading Instance Feature