Ahmad Esmaeili
Computer and Information Technology, Purdue University.
Knoy 411, 401 N. Grant St.
West Lafayette, IN 47907
I am a graduate assistant in the department of Computer and Information Technology at Purdue University, working with Prof. Eric Matson. My research interests are Distributed Artificial Intelligence, Multi-agent Systems, and Collaborative Machine Learning. Currently, my research project focuses on agent-based distributed learning and the design of decentralized multi-agent based machine learning system.
news
| Dec 20, 2023 | Our paper on Holonic Learning has been accepted in the main track of AAMAS 24. |
|---|---|
| Nov 9, 2023 | A paper accepted in the IEEE Robotoc Computing. |
| Nov 6, 2023 | Put a preprint on arXiv (Distributed Agent-Based Collaborative Learning in Cross-Individual Wearable Sensor-Based Human Activity Recognition: arXiv:2311.04236). |
| Sep 12, 2023 | Put a preprint on arXiv (Hybrid Algorithm Selection and Hyperparameter Tuning on Distributed Machine Learning Resources: A Hierarchical Agent-based Approach: arXiv:2309.06604). |
| May 5, 2023 | A paper accepted in the journal of Systems. |
highlighted projects
Holonic Learning
A research on designing a collaborative and privacy-focused framework for training deep learning models, leveraging structured self-similar hierarchies and individual model aggregation within holons to address scalability, resource distribution, and privacy concerns in the context of increasingly distributed machine learning paradigms.
Distributed Cross-Individual Human Activity Recognition
A research on a collaborative distributed learning approach rooted in multi-agent principles for decentralized Human Activity Recognition, leveraging wearable sensor technologies to uphold privacy, eliminate external server dependencies, and demonstrate superior effectiveness in local and global generalization.
Agent-based Distributed ML Algorithm Selection and Tuning
A research on developing a fully automated and collaborative agent-based mechanism for ML algorithm selection and hyperparameter tuning, utilizing resources organized distributedly by a hierarchical machine-learning platform.
Agent-based Modelling of Distributed Machine Learning Systems
A research on building a hybrid machine learning platform that leverages Multi-Agent Systems to autonomously organize and democratize geographically distributed ML resources and offers analytical capabilities for robust research assessment across various algorithms and datasets.