Supervised Learning with Dynamic Network Architectures

Carlström, Herman; Slottner Seholm, Filip

Supervised Learning with Dynamic Network Architectures

dc.contributor.author	Carlström, Herman
dc.contributor.author	Slottner Seholm, Filip
dc.contributor.department	Chalmers tekniska högskola / Institutionen för data och informationsteknik	sv
dc.contributor.examiner	Haghir Chehreghani, Morteza
dc.contributor.supervisor	Strannegård, Claes
dc.date.accessioned	2019-07-19T09:25:44Z
dc.date.available	2019-07-19T09:25:44Z
dc.date.issued	2019	sv
dc.date.submitted	2019
dc.description.abstract	There exists many different techniques for training neural networks, but few are designed to handle the case of lifelong learning. Most models are based on the assumption that there is a training phase with a finite amount of data. This thesis investigates and evaluates a brand new algorithm, namely the Lifelong Learning starting from zero (LL0) which can be used for lifelong learning where there is a continuous stream of data. The algorithm stems from biology, logical rules, and machine learning. The algorithm builds a dynamic artificial neural network architecture over time, based on four different concepts; extension, generalization, forgetting and backpropagation. These first three concepts all have their origin in biology, and can be found in animals and humans in the form of neuroplasticity. The model is evaluated and benchmarked against five other models on different datasets and problems. The obtained results act as a proof of concept for the algorithm. Lastly, the pros and cons of the model are discussed, followed by a discussion on future work. The model proposed outperforms all models on chosen benchmarks, primarily in the area of one-shot learning. The model manages to achieve about 90% accuracy on unseen data after only training on a small portion of the training set on multiple datasets. LL0 also shows promising results in the area of lifelong learning.	sv
dc.identifier.coursecode	DATX05	sv
dc.identifier.uri	https://hdl.handle.net/20.500.12380/300060
dc.language.iso	eng	sv
dc.setspec.uppsok	Technology
dc.subject	Machine Learning	sv
dc.subject	Neural Networks	sv
dc.subject	Dynamic Architectures	sv
dc.subject	Supervised Learning	sv
dc.subject	Lifelong Learning	sv
dc.subject	One-Shot Learning	sv
dc.subject	Transfer Learning	sv
dc.subject	Computer Science	sv
dc.title	Supervised Learning with Dynamic Network Architectures	sv
dc.type.degree	Examensarbete för masterexamen	sv
dc.type.uppsok	H

Ladda ner

Original bundle

Visar 1 - 1 av 1

Namn:: CSE 19-57 Carlstrom_Slottner Seholm.pdf
Storlek:: 27.51 MB
Format:: Adobe Portable Document Format
Beskrivning:: Supervised Learning With Dynamic Network Architectures

Ladda ner

License bundle

Visar 1 - 1 av 1

Namn:: license.txt
Storlek:: 1.14 KB
Format:: Item-specific license agreed upon to submission
Beskrivning:

Ladda ner

Samling

Examensarbeten för masterexamen