N/A

Teaching Experience Publication ML ∪ Systems Resources Contact
2025
C15
Variant Generation and Diversity Strategies for Enhanced Variant Parallelism in Edge AI
Navidreza Asadi, Sagnik Dutta, Wolfgang Kellerer

IEEE Globecom 2025 Workshop on A4E: AI/ML for Edge/Fog Networks (accepted)

Abstract

C14
Comparative Performance and Cost Analysis of Computer Vision in Edge-Cloud Continuum
Hongyu Guo, Navidreza Asadi, Giovanni Bartolomeo, Patrick Laufer, Jörg Ott, Wolfgang Kellerer

ACM CoNEXT ECCAI 2025 (accepted)

Abstract

C13
Road to Tiny Reality: Digital Twins for Decentralized AI on Microcontrollers
Navidreza Asadi*, Lars Wulfert*, Halil İbrahim Bengu*, Hendrik Wöhrle, Wolfgang Kellerer

ACM MobiCom 2025 (accepted)

* Equal Contribution Abstract

C12
Gist - Optimizing Segmentation for Decentralized Federated Learning on Tiny Devices
Navidreza Asadi*, Halil İbrahim Bengu*, Lars Wulfert*, Hendrik Wöhrle, Wolfgang Kellerer

ACM MobiCom FLEdge-AI 2025 (accepted)

* Equal Contribution Abstract

C11
Challenges in Designing Robust RL-Based Autoscalers
Navidreza Asadi*, Dalal Ali*, Răzvan-Mihai Ursu, Wolfgang Kellerer

ACM SOSP 2025 Practical Adoption Challenges of ML for Systems Workshop (SOSP-PACMI '25) (accepted)

* Equal Contribution Abstract

C10
WaveSurfer - Scheduling Irregular Pulsing Attacks on Microservice Autoscaling
Navidreza Asadi, Răzvan-Mihai Ursu, Johannes Zerwas, Leon Wong, Wolfgang Kellerer

ACM SIGCOMM 2025 Posters and Demos

Abstract

C9
Simurgh: Multi-Agent Adversarial Benchmarking for Proactive Microservice Observability
Navidreza Asadi*, Răzvan-Mihai Ursu*, Leon Wong, Wolfgang Kellerer

ACM SIGCOMM 2025 Next-Generation Network Observability Workshop (SIGCOMM-NGNO 2025)

* Equal Contribution Acceptance Rate: 33% (lowest among all workshops at SIGCOMM 2025) Abstract

C8
On-Demand Container Partitioning for Distributed ML
Giovanni Bartolomeo*, Navidreza Asadi*, Wolfgang Kellerer, Jörg Ott, Nitinder Mohan

USENIX Annual Technical Conference (USENIX ATC 2025)

* Equal Contribution Acceptance Rate: 15% Abstract

C7
Comparative Analysis Between Decentralized and Centralized Network Digital Twins of Kubernetes Clusters
Răzvan-Mihai Ursu, Navidreza Asadi, Johannes Zerwas, Leon Wong, Wolfgang Kellerer

11th IEEE International Conference on Network Softwarization (NetSoft 2025)

Abstract

C6
Beyond Layers: Container Registries for Files Distribution and On-Demand Image Partitioning
Giovanni Bartolomeo*, Navidreza Asadi*, Wolfgang Kellerer, Jörg Ott, Nitinder Mohan

ACM EuroSys 2025 Best Poster Award

* Equal Contribution Abstract

2024
C5
Towards Data-Driven Approaches for Network Digital Twins of Microservice-Based Architectures
Răzvan-Mihai Ursu, Navidreza Asadi, Leon Wong, Wolfgang Kellerer

WueWoWas '24 Best Paper Award

Abstract

J2
FullPack: Full Vector Utilization for Sub-Byte Quantized Matrix-Vector Multiplication on General Purpose CPUs
Hossein Katebi, Navidreza Asadi, Maziar Goudarzi

IEEE Computer Architecture Letters (CAL)

Abstract

J1
Variant Parallelism: Lightweight Deep Convolutional Models for Distributed Inference on IoT Devices
Navidreza Asadi, Maziar Goudarzi

IEEE Internet of Things Journal Impact Factor: 10.6

Abstract

2023
C4
Adaptive Decentralized Federated Gossip Learning for Resource-Constrained IoT Devices
Lars Wulfert, Navidreza Asadi, Wen-Yu Chung, Christian Wiede, Anton Grabmaier

ACM CoNEXT Distributed ML Workshop 2023

Acceptance Rate: 35% Abstract

C3
HYPA: Hybrid Horizontal Pod Autoscaling with Automated Model Updates
Kaan Aykurt, Răzvan-Mihai Ursu, Johannes Zerwas, Patrick Krämer, Navidreza Asadi, Leon Wong, Wolfgang Kellerer

IEEE Conference on Network Function Virtualization and Software Defined Networks

Abstract

C2
Towards Digital Network Twins: Can We Machine Learn Network Function Behaviors?
Răzvan-Mihai Ursu, Johannes Zerwas, Patrick Krämer, Navidreza Asadi, Phil Rodgers, Leon Wong, Wolfgang Kellerer

IEEE 9th International Conference on Network Softwarization (NetSoft)

Abstract

2022
C1
KAPETÁNIOS: Automated Kubernetes Adaptation through a Digital Twin
Johannes Zerwas, Patrick Krämer, Răzvan-Mihai Ursu, Navidreza Asadi, Phil Rodgers, Leon Wong, Wolfgang Kellerer

13th International Conference on Network of the Future (NoF)

Abstract

R2
FullPack: Full Vector Utilization for Sub-Byte Quantized Inference on General Purpose CPUs
Hossein Katebi, Navidreza Asadi, Maziar Goudarzi

Full paper version. A shorter version is accepted at IEEE CAL. Abstract

R1
An Ensemble Mobile-Cloud Computing Method for Affordable and Accurate Glucometer Readout
Navidreza Asadi, Maziar Goudarzi

Abstract

Contact


Room 1912, Chair of Communication Networks, CE Department, Technical University of Munich, Munich, Germany.