SNAI 2024 invited original research contributions on the two-way intersection of computer networking and artificial intelligence. Papers were reviewed double-blind by at least three independent reviewers, with author rebuttal and a single-track oral programme plus a poster session.

Important dates

Workshop proposals

15 March 2024

Paper submission deadline

28 June 2024 (23:59 AoE)

Acceptance notification

16 August 2024

Camera-ready due

13 September 2024

Conference

18–20 October 2024

Submission categories

• Full papers (8 pages + references): original research contributions with comprehensive evaluation.
• Short papers (4 pages + references): preliminary results, system experiences, position papers, or negative results.
• Poster papers (2 pages): late-breaking results or work in progress, presented at the poster session.

Track 1 — AI for Network Management and Optimization

Learned congestion control, traffic engineering, routing, scheduling, network slicing, intent-based and self-driving networks, closed-loop network automation, telemetry summarization, configuration synthesis, and operator-facing AI tools. Submissions should demonstrate measurable improvement on representative networking workloads.

Track 2 — Edge and In-Network Inference

Model placement and partitioning across the edge and the data plane, in-switch and SmartNIC inference, line-rate feature extraction, latency-aware serving, and hardware-software co-design for in-network ML. Work combining systems measurement with realistic model evaluation was particularly welcome.

Track 3 — Distributed and Federated Learning over Networks

Communication-efficient and privacy-preserving training, gradient compression and quantization, federated learning under heterogeneous and bandwidth-constrained links, straggler mitigation, and decentralized training over peer-to-peer topologies. Submissions were expected to include a realistic network model.

Track 4 — Machine Learning for Network Security

Intrusion and anomaly detection, encrypted-traffic classification, attack attribution, evasion and adversarial robustness, large-scale measurements of malicious traffic, and the operational deployment of ML-based detection in production networks.

Review process

All papers were reviewed double-blind. Authors were required to anonymize their submission, including references phrased in the third person. Each paper received at least three independent reviews. The program chairs resolved disagreements and consulted additional reviewers for borderline cases. The final acceptance rate was 27.6% (24 papers accepted from 87 submissions).

Publication and indexing

Accepted papers were published in the open-access SNAI 2024 proceedings, archived in the LiteConf repository, and assigned DOIs. Authors retained copyright. The proceedings were submitted to major indexing services.

Double-blind policies

• Remove author names, affiliations, and acknowledgments from the submitted PDF.
• Phrase self-citations in the third person ("Prior work [12] showed...") rather than "Our prior work [12]...".
• Do not include hyperlinks that reveal identity (e.g. personal webpages, GitHub organisations named after the author).
• Pre-prints already on arXiv were permitted but should be cited as a third-party work.

Conflicts of interest

Authors were asked to declare conflicts of interest for the program committee during submission. A conflict existed when an author was at the same institution as a potential reviewer, had co-authored with them in the past two years, or had a personal relationship.

Contact

Questions about the archived call may be sent to [email protected]. For the current edition, see the SNAI 2026 call for papers.