ISBNet: Instance-aware Selective Branching Network

05/13/2019
by   Shaofeng Cai, et al.
0

Recent years have witnessed growing interests in designing efficient neural networks and neural architecture search (NAS). Although remarkable efficiency and accuracy have been achieved, existing expert designed and NAS models neglect that input instances are of varying complexity thus different amount of computation is required. Therefore, inference with a fixed model that processes all instances through the same transformations would waste plenty of computational resources. Customizing the model capacity in an instance-aware manner is highly demanded. In this paper, we introduce a novel network ISBNet to address this issue, which supports efficient instance-level inference by selectively bypassing transformation branches of infinitesimal importance weight. We also propose lightweight hypernetworks SelectionNet to generate these importance weights instance-wisely. Extensive experiments have been conducted to evaluate the efficiency of ISBNet and the results show that ISBNet achieves extremely efficient inference comparing to existing networks. For example, ISBNet takes only 12.45 state-of-the-art efficient network ShuffleNetV2 with comparable accuracy.

READ FULL TEXT
research
11/26/2018

InstaNAS: Instance-aware Neural Architecture Search

Neural Architecture Search (NAS) aims at finding one "single" architectu...
research
11/02/2020

PV-NAS: Practical Neural Architecture Search for Video Recognition

Recently, deep learning has been utilized to solve video recognition pro...
research
08/30/2022

You Only Search Once: On Lightweight Differentiable Architecture Search for Resource-Constrained Embedded Platforms

Benefiting from the search efficiency, differentiable neural architectur...
research
12/24/2019

BETANAS: BalancEd TrAining and selective drop for Neural Architecture Search

Automatic neural architecture search techniques are becoming increasingl...
research
09/03/2019

MANAS: Multi-Agent Neural Architecture Search

The Neural Architecture Search (NAS) problem is typically formulated as ...
research
03/17/2021

The Untapped Potential of Off-the-Shelf Convolutional Neural Networks

Over recent years, a myriad of novel convolutional network architectures...
research
10/02/2020

Dynamic Graph: Learning Instance-aware Connectivity for Neural Networks

One practice of employing deep neural networks is to apply the same arch...

Please sign up or login with your details

Forgot password? Click here to reset