LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes

by   Aditya Kusupati, et al.

Learning binary representations of instances and classes is a classical problem with several high potential applications. In modern settings, the compression of high-dimensional neural representations to low-dimensional binary codes is a challenging task and often require large bit-codes to be accurate. In this work, we propose a novel method for Learning Low-dimensional binary Codes (LLC) for instances as well as classes. Our method does not require any side-information, like annotated attributes or label meta-data, and learns extremely low-dimensional binary codes ( 20 bits for ImageNet-1K). The learnt codes are super-efficient while still ensuring nearly optimal classification accuracy for ResNet50 on ImageNet-1K. We demonstrate that the learnt codes capture intrinsically important features in the data, by discovering an intuitive taxonomy over classes. We further quantitatively measure the quality of our codes by applying it to the efficient image retrieval as well as out-of-distribution (OOD) detection problems. For ImageNet-100 retrieval problem, our learnt binary codes outperform 16 bit HashNet using only 10 bits and also are as accurate as 10 dimensional real representations. Finally, our learnt binary codes can perform OOD detection, out-of-the-box, as accurately as a baseline that needs  3000 samples to tune its threshold, while we require none. Code and pre-trained models are available at


page 16

page 17


Efficient Passage Retrieval with Hashing for Open-domain Question Answering

Most state-of-the-art open-domain question answering systems use a neura...

Asymmetric Hash Code Learning for Remote Sensing Image Retrieval

Remote sensing image retrieval (RSIR), aiming at searching for a set of ...

Binary Generative Adversarial Networks for Image Retrieval

The most striking successes in image retrieval using deep hashing have m...

Deep Learning to Ternary Hash Codes by Continuation

Recently, it has been observed that 0,1,-1-ternary codes which are simpl...

Binary Paragraph Vectors

Recently Le & Mikolov described two log-linear models, called Paragraph ...

Compact Compositional Models

Learning compact and interpretable representations is a very natural tas...

MixFaceNets: Extremely Efficient Face Recognition Networks

In this paper, we present a set of extremely efficient and high throughp...