Certifying Stability and Performance of Uncertain Differential-Algebraic Systems: A Dissipativity Framework

no code implementations • 16 Aug 2023 • Emily Jensen, Neelay Junnarkar, Murat Arcak, Xiaofan Wu, Suat Gumussoy

This paper presents a novel framework for characterizing dissipativity of uncertain dynamical systems subject to algebraic constraints.

PD-MORL: Preference-Driven Multi-Objective Reinforcement Learning Algorithm

1 code implementation • 16 Aug 2022 • Toygun Basaklar, Suat Gumussoy, Umit Y. Ogras

To this end, we propose a novel MORL algorithm that trains a single universal network to cover the entire preference space scalable to continuous robotic tasks.

Continuous Control Multi-Objective Reinforcement Learning +1

Hypervector Design for Efficient Hyperdimensional Computing on Edge Devices

no code implementations • 8 Mar 2021 • Toygun Basaklar, Yigit Tuncel, Shruti Yadav Narayana, Suat Gumussoy, Umit Y. Ogras

The proposed approach decreases the hypervector dimension by more than $32\times$ while maintaining or increasing the accuracy achieved by conventional HDC.

Transfer Learning for Human Activity Recognition using Representational Analysis of Neural Networks

no code implementations • 5 Dec 2020 • Sizhe An, Ganapati Bhat, Suat Gumussoy, Umit Ogras

The typical approach is training a HAR classifier offline with known users and then using the same classifier for new users.

Human Activity Recognition Transfer Learning