An Investigation of the Weight Space for Version Control of Neural Networks

06/18/2020
by   Konstantin Schürholt, et al.
0

Deployed Deep Neural Networks (DNNs) are often trained further to improve in performance. This complicates the tracking of DNN model versions and the synchronization between already deployed models and upstream updates of the same architecture. Software Version Control cannot be applied straight-forwardly to DNNs due to the different nature of software and DNN models. In this paper we investigate if the weight space of DNN models contains a structure, which can be used for the identification of individual DNN models. Our results show that DNN models evolve on unique, smooth trajectories in weight space which we can exploit as feature for DNN version control.

READ FULL TEXT
research
01/09/2021

SyReNN: A Tool for Analyzing Deep Neural Networks

Deep Neural Networks (DNNs) are rapidly gaining popularity in a variety ...
research
04/10/2021

Use of Metamorphic Relations as Knowledge Carriers to Train Deep Neural Networks

Training multiple-layered deep neural networks (DNNs) is difficult. The ...
research
06/06/2020

Guarded Deep Learning using Scenario-Based Modeling

Deep neural networks (DNNs) are becoming prevalent, often outperforming ...
research
10/12/2021

Zero-bias Deep Neural Network for Quickest RF Signal Surveillance

The Internet of Things (IoT) is reshaping modern society by allowing a d...
research
10/06/2021

Generalizing Neural Networks by Reflecting Deviating Data in Production

Trained with a sufficiently large training and testing dataset, Deep Neu...
research
11/18/2022

A Tale of Two Cities: Data and Configuration Variances in Robust Deep Learning

Deep neural networks (DNNs), are widely used in many industries such as ...
research
06/16/2023

Representation and decomposition of functions in DAG-DNNs and structural network pruning

The conclusions provided by deep neural networks (DNNs) must be carefull...

Please sign up or login with your details

Forgot password? Click here to reset