Budget-Aware Adapters for Multi-Domain Learning

05/15/2019
by   Rodrigo Berriel, et al.
0

Multi-Domain Learning (MDL) refers to the problem of learning a set of models derived from a common deep architecture, each one specialized to perform a task in a certain domain (e.g., photos, sketches, paintings). This paper tackles MDL with a particular interest in obtaining domain-specific models with an adjustable budget in terms of the number of network parameters and computational complexity. Our intuition is that, as in real applications the number of domains and tasks can be very large, an effective MDL approach should not only focus on accuracy but also on having as few parameters as possible. To implement this idea we derive specialized deep models for each domain by adapting a pre-trained architecture but, differently from other methods, we propose a novel strategy to automatically adjust the computational complexity of the network. To this aim, we introduce Budget-Aware Adapters that select the most relevant feature channels to better handle data from a novel domain. Some constraints on the number of active switches are imposed in order to obtain a network respecting the desired complexity budget. Experimentally, we show that our approach leads to recognition accuracy competitive with state-of-the-art approaches but with much lighter networks both in terms of storage and computation.

READ FULL TEXT

page 2

page 10

research
09/20/2023

Budget-Aware Pruning: Handling Multiple Domains with Less Parameters

Deep learning has achieved state-of-the-art performance on several compu...
research
11/02/2020

Searching k-Optimal Goals for an Orienteering Problem on a Specialized Graph with Budget Constraints

We propose a novel non-randomized anytime orienteering algorithm for fin...
research
10/14/2022

Parameter Sharing in Budget-Aware Adapters for Multi-Domain Learning

Deep learning has achieved state-of-the-art performance on several compu...
research
03/25/2020

Not all domains are equally complex: Adaptive Multi-Domain Learning

Deep learning approaches are highly specialized and require training sep...
research
11/21/2017

Residual Parameter Transfer for Deep Domain Adaptation

The goal of Deep Domain Adaptation is to make it possible to use Deep Ne...
research
07/17/2023

Multi-Domain Learning with Modulation Adapters

Deep convolutional networks are ubiquitous in computer vision, due to th...
research
04/09/2019

Deep Virtual Networks for Memory Efficient Inference of Multiple Tasks

Deep networks consume a large amount of memory by their nature. A natura...

Please sign up or login with your details

Forgot password? Click here to reset