İlkay Ulusoy

E-mail

ilkay@metu.edu.tr

Department

Department of Electrical and Electronics Engineering

ORCID

0000-0002-0863-7116

Scopus Author ID

8707741300

Web of Science Researcher ID

AAZ-9415-2020

Analysis of Model-Agnostic Meta-Reinforcement Learning on Automated HVAC Control Filiz, Ulas; Hekimoglu, Mehmet Burak; Alioğlu, Alper; Ulusoy, İlkay (2025-01-01) This paper introduces a Model-Agnostic Meta-Reinforcement Learning framework for HVAC automation, integrating Model Agnostic Meta-Learning with Double Deep Q-Networks to improve adaptability across varying environmental co...
Benchmarking Deep Learning Models For Automated Waste Classification Donmez, Mert; Alioğlu, Alper; Ulusoy, İlkay (2025-01-01) Recycling is essential for reducing waste, conserving resources, and promoting a more sustainable and efficient use of materials. However, manual sorting is inefficient, prone to human error, and limits recycling efficienc...
Accelerating Reinforcement Learning for HVAC Systems Using an LSTM-based Surrogate Simulator Hekimoglu, Mehmet Burak; Alioğlu, Alper; Filiz, Ulas; Ulusoy, İlkay; Schmidt, Klaus Verner (2025-01-01) Reinforcement learning (RL) has shown great potential in optimizing the operation of HVAC systems, improving energy efficiency, and enhancing user comfort. However, the slow input/output operations associated with simulati...
Material Characterization Based on Capacitive and Dielectric Properties for Waste Recycling Adnan, Ashhar; Alioğlu, Alper; Ulusoy, İlkay (2025-01-01) This paper outlines the development process for a capacitive sensing technique used to detect the material of an object placed in the sensor field. The sensor operates by detecting small capacitive changes due to the varia...
IoT-Based Multi-Speed HVAC Control: Simulation Results and Testbed Development Hekimoglu, Mehmet Burak; Alioğlu, Alper; Buyuker, Asiye Irem; Ulusoy, İlkay (2025-01-01) This paper presents a multi-speed rule-based controller for HVAC and ventilation management, evaluated in simulation and prepared for real-world deployment. The controller uses real-time environmental data and predefined t...
Ontology-Driven Smart Building Semantics: An EnergyPlus-SAREF Approach Kirnapci, Utku; Hekimoglu, Mehmet Burak; Alioğlu, Alper; Ulusoy, İlkay (2025-01-01) Semantic ontologies like SAREF provide a standardized framework for interoperability in smart buildings. However, its potential for metadata extraction remains underexplored. This work introduces a novel approach that util...
Improved brain effective connectivity modelling by dynamic Bayesian networks Ulusoy, İlkay; Geduk, Salih (2024-09-01) Background: If brain effective connectivity network modelling (ECN) could be accurately achieved, early diagnosis of neurodegenerative diseases would be possible. It has been observed in the literature that Dynamic Bayesia...
Gradient-descent hardware-aware training and deployment for mixed-signal neuromorphic processors Cakal, Ugurcan; Maryada, Maryada; Wu, Chenxi; Ulusoy, İlkay; Muir, Dylan Richard (2024-03-01) Mixed-signal neuromorphic processors provide extremely low-power operation for edge inference workloads, taking advantage of sparse asynchronous computation within spiking neural networks (SNNs). However, deploying robust ...
Elimination of Non-Novel Segments at Multi-Scale for Few-Shot Segmentation Kayabasi, Alper; Tufekci, Gulin; Ulusoy, İlkay (2023-01-01) Few-shot segmentation aims to devise a generalizing model that segments query images from unseen classes during training with the guidance of a few support images whose class tally with the class of the query. There exist ...
Detecting Driver Drowsiness as an Anomaly Using LSTM Autoencoders Tüfekci, Gülin; Kayabaşı, Alper; Akagündüz, Erdem; Ulusoy, İlkay (2023-01-01) In this paper, an LSTM autoencoder-based architecture is utilized for drowsiness detection with ResNet-34 as feature extractor. The problem is considered as anomaly detection for a single subject; therefore, only the norma...

Citation Formats