Erich Malan

Dottorando in Ingegneria Informatica E Dei Sistemi , 38^o ciclo (2022-2025)
Dipartimento di Automatica e Informatica (DAUIN)

Docente esterno e/o collaboratore didattico
Dipartimento di Automatica e Informatica (DAUIN)

erich.malan@polito.it

Profilo

Dottorato di ricerca

Argomento di ricerca

Distributed and Federated Learning over IoT networks

Tutori

Presentazione della ricerca

Poster

Interessi di ricerca

Computer architectures and Computer aided design

Biografia

Erich Malan earned an M.Sc. in Data Science and Engineering from the Politecnico di Torino in 2021. He is pursuing a PhD at the Electronic Design Automation (EDA) Group within the Department of Control and Computer Engineering at the same institution. His research investigates deep learning methods for processing the large volume of data gathered by ubiquitous IoT networks. Centralizing such data faces is constrained by privacy regulations, wireless network bottlenecks, and data management costs. Distributed and federated learning data management paradigms offer an alternative avenue for training deep learning models over decentralized data at the edge of the IoT. He is particularly interested in optimizing federated learning strategies to minimize communication overhead and effectively manage the limited resources of low-power IoT nodes. Additionally, his research extends to the accompanying challenges of data and system heterogeneity, automating data annotation, and security and privacy threats. The ultimate goal is to promote the widespread deployment of private and efficient intelligent services across IoT environments.

Premi e riconoscimenti

Certificate of attendance to conference as a speaker at the IEEE International Symposium on Low Power Electronics and Design (ISLPED) 2023. (2023)

Didattica

Insegnamenti

Corso di laurea di 1° livello

Informatica. A.A. 2024/25, INGEGNERIA AEROSPAZIALE. Collaboratore del corso
Informatica. A.A. 2023/24, INGEGNERIA INFORMATICA. Collaboratore del corso
Informatica. A.A. 2022/23, INGEGNERIA INFORMATICA. Collaboratore del corso

MostraNascondi A.A. passati

Ricerca

Gruppi di ricerca

EDA - Electronic Design Automation (DAUIN)

Pubblicazioni

Pubblicazioni più recenti Vedi tutte le pubblicazioni su Porto@Iris

Malan, Erich; Peluso, Valentino; Calimera, Andrea; Macii, Enrico (In stampa)
Draft & Refine: Efficient Resource Management in Federated Learning under Pathological Labels Skew. In: International Conference on Electronics Circuits and Systems (ICECS) 2024
Contributo in Atti di Convegno (Proceeding)
Peluso, Valentino; Malan, Erich; Calimera, Andrea; Macii, Enrico (In stampa)
Private Tensor Freezing for an Efficient Federated Learning with Homomorphic Encryption. In: International Conference on Computer Design (ICCD) 2024
Contributo in Atti di Convegno (Proceeding)
Malan, Erich; Peluso, Valentino; Calimera, Andrea; Macii, Enrico; Montuschi, Paolo (2024)
Automatic Layer Freezing for Communication Efficiency in Cross-Device Federated Learning. In: IEEE INTERNET OF THINGS JOURNAL, vol. 11, pp. 6072-6083. ISSN 2327-4662
Contributo su Rivista
Malan, Erich; Peluso, Valentino; Calimera, Andrea; Macii, Enrico (2023)
Communication-Efficient Federated Learning with Gradual Layer Freezing. In: IEEE EMBEDDED SYSTEMS LETTERS, vol. 15, pp. 25-28. ISSN 1943-0663
Contributo su Rivista
Malan, Erich; Peluso, Valentino; Calimera, Andrea; Macii, Enrico (2023)
Enabling DVFS Side-Channel Attacks for Neural Network Fingerprinting in Edge Inference Services. In: International Symposium on Low Power Electronics and Design, Vienna (AUT), 07-08 August 2023, pp. 1-6. ISBN: 979-8-3503-1175-4
Contributo in Atti di Convegno (Proceeding)