Meistaravörn í tölvunarfræði: Frímann Freyr Kjérúlf Björnsson

Evaluating Transferability of Non-Local Features in Deep Neural Networks

  • 3.6.2019, 13:00 - 14:00

Student: Frímann Freyr Kjérúlf Björnsson
Title: Evaluating Transferability of Non-Local Features in Deep Neural Networks
Date and Location: June 3rd 2019 at 13:00 in room V102
Supervisor: Yngvi Björnsson, professor Department of Computer Science

Abstract: Transfer learning is becoming an essential part of modern machine learning, especially in the field of deep neural networks. In the domain of image recognition there are known methods for evaluating to what degree a feature extractor can be considered general to the domain, or specific to the task at hand. The method mainly consists of measuring transferability of a neural network as a function of number of transferred layers. This is of high importance when aiming for a successful knowledge transfer since one typically wants to transfer only the general feature extractors and leave the specific ones behind. The general features in the case of image classification can be considered local with respect to each pixel, since the feature extractors in early layers activate on simple features like edges, which are localized within a given radius from a given pixel. One might then ask the question, whether similar methods are also applicable in domains characterized by non-local features. Chess is as excellent example of such a domain since a square's locality can not be defined by the adjacent squares alone. One needs to take into account that a single piece can traverse the whole board in a single move. We show that this method is applicable in the case of chess endgame tablebases, in spite of structural differences in the feature space, and that the knowledge distribution has a similar structure as in the case of image classification.



Vinsamlegast athugið að á viðburðum Háskólans í Reykjavík (HR) eru teknar ljósmyndir og myndbönd sem notuð eru í markaðsstarfi HR. Hægt er að nálgast frekari upplýsingar á ru.is eða með því að senda tölvupóst á netfangið: personuvernd@ru.is
//
Please note that at events hosted at Reykjavik University (RU), photographs and videos are taken which might be used for RU marketing purposes. Read more about this on our ru.is or send an e-mail: personuvernd@ru.is