Pith. sign in

REVIEW

Model Partition and Resource Allocation for Split Learning in Vehicular Edge Networks

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2411.06773 v1 pith:AZOK7MA3 submitted 2024-11-11 cs.LG cs.DC

Model Partition and Resource Allocation for Split Learning in Vehicular Edge Networks

classification cs.LG cs.DC
keywords learningsplitvehicularallocationcommunicationdatanetworksresource
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

The integration of autonomous driving technologies with vehicular networks presents significant challenges in privacy preservation, communication efficiency, and resource allocation. This paper proposes a novel U-shaped split federated learning (U-SFL) framework to address these challenges on the way of realizing in vehicular edge networks. U-SFL is able to enhance privacy protection by keeping both raw data and labels on the vehicular user (VU) side while enabling parallel processing across multiple vehicles. To optimize communication efficiency, we introduce a semantic-aware auto-encoder (SAE) that significantly reduces the dimensionality of transmitted data while preserving essential semantic information. Furthermore, we develop a deep reinforcement learning (DRL) based algorithm to solve the NP-hard problem of dynamic resource allocation and split point selection. Our comprehensive evaluation demonstrates that U-SFL achieves comparable classification performance to traditional split learning (SL) while substantially reducing data transmission volume and communication latency. The proposed DRL-based optimization algorithm shows good convergence in balancing latency, energy consumption, and learning performance.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.