The introduction of virtualization, softwarization, and programmability in 5G has reshaped transport networks into dynamic, software-defined systems that can meet the demands of AI workloads. In this evolving paradigm, the transport layer becomes a key enabler—moving from passive conduit to intelligent, programmable substrate for distributed AI.
As inference and training shift toward the edge, transport systems must ensure not only bandwidth and reliability but also adaptability, low-latency in-network processing, and cross-layer integration—especially across heterogeneous wireless and optical domains.
TRAIN 2025 provides a focused venue to explore how transport architectures must evolve to support AI-native operations. The workshop emphasizes co-design across physical-layer programmability (e.g., P4, eBPF), AI algorithms, and orchestration strategies, supporting applications such as Federated Learning, autonomous systems, and real-time inference
We invite contributions on topics including in-network compute, SDN-based orchestration, telemetry, and AI-native transport mechanisms.
Submission Guidelines: Submitted manuscripts must follow the IEEE 2-column conference style and are limited to 6 pages including references. The submission and review process will be handled through a dedicated track on the EDAS platform.
🔗 Submission Link: https://edas.info/N34170
Topics of Interest
AI-driven transport protocols and architectures
Programmable transport for AI workloads
In-network aggregation and distributed AI
Federated/edge learning over SDN transport
Transport-aware orchestration of AI pipelines
Telemetry-driven transport optimization
High-performance model delivery/inference
Physical-layer co-design (optical, mmWave)
Experimental testbeds for AI-native transport
Energy-efficient AI transport at the edge
Transport in 6G RAN/core network AI
DPUs and SmartNICs for transport acceleration
Multi-generation optical network design
Digital twin-based network management
Network resilience and survivability
Optical backhaul/midhaul/fronthaul in 6G
Open/disaggregated transport architectures
Paper Submission Deadline: August 11, 2025 August 25, 2025 August 31, 2025
Notification of Acceptance: September 8, 2025
Camera-ready Submission: September 14, 2025
Workshop Date: TBD
Isabella Cerutti, Researcher at Joint Research Centre
Nicola Di Cicco, Operations Research Specialist at Optit S.r.l
Alessandro Pacini, Telecom Cloud Engineer at BubbleRAN
Davide Scano, R&D Engineer at Nokia
Koteswararao Kondepu, Associate Professor at Indian Institute of Technology (Dharwad)
Emilio Paolini, General Co-Chair, Scuola Superiore Sant’Anna, Pisa, Italy
Mëmëdhe Ibrahimi, General Co-Chair, Politecnico di Milano, Milan, Italy
Gianluca Davoli, TPC Chair, Università di Bologna, Bologna, Italy