Master thesis defense by Aliya Mohammed

Master thesis defense by Aliya Mohammed

Johannes Rabold發表於
Number of replies: 0

Dear all,

you are cordially invited to join the master thesis defense of Aliya Mohammed. It takes place in a hybrid fashion on Monday, 27th of June, 13:00 in the CogSys Lab (WE5.05.013) or via Teams (https://teams.microsoft.com/l/meetup-join/19%3ameeting_ZjQ3MjBlNzMtMTFmYS00OGM2LTg4MmItMzAxZjAzMzQ2NDRk%40thread.v2/0?context=%7b%22Tid%22%3a%224f18ddfc-c31f-4597-afda-fa5a760bf3cf%22%2c%22Oid%22%3a%2243265e8c-94ad-4d4d-946e-d080c67e8953%22%7d).

Everyone is invited to join. Title and Abstract can be found below.

Best
Johannes

Title: Optimizing Hierarchical Classification Problems for Resource-Constrained Systems Using Reinforcement Learning Methods

Abstract:
In predictive maintenance, machine learning models are often deployed at the edge
on resource-constrained embedded devices. Approximating machine learning models
to deploy them at the edge might compromise the predictive accuracy. On the other
hand, deploying complex machine learning models at the edge might cause hardware
overheating and decrease the life span of the edge devices. In [1 ], the authors propose
hierarchical classification to achieve a trade-off between predictive accuracy and energy consumption at inference. The authors design a reinforcement learning framework to select the optimal classifier at every level of the hierarchical classification problem. However, the authors only consider the estimation of the energy. We introduce the
HierarchicalEdge framework that extends the previous method by considering real
energy measurements. In addition, the HierarchicalEdge framework implements feature selection, as well as model selection. The classifiers selected by the HierarchicalEdge framework consume low energy at inference while maintaining high predictive accuracy.
This thesis shows that the HierarchicalEdge framework is superior to both flat classifiers
and the hierarchical classifier selected by the reinforcement learning framework of [1 ].
Additionally, the thesis shows that the classifiers and feature sets selected by the
HierarchicalEdge framework showed better accuracy and energy consumption trade-off
than both flat classifiers with simple features and flat classifiers with complex features.

[1] Stephen Adams, Ryan Meekins, Peter A Beling, Kevin Farinholt, Nathan Brown,
Sherwood Polter, and Qing Dong. Hierarchical fault classification for resource
constrained systems. Mechanical Systems and Signal Processing, 134:106–266,
2019