{"ID":2844483,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.06563","arxiv_id":"2511.06563","title":"Practical Policy Distillation for Reinforcement Learning in Radio Access Networks","abstract":"Adopting artificial intelligence (AI) in radio access networks (RANs) presents several challenges, including limited availability of link-level measurements (e.g., CQI reports), stringent real-time processing constraints (e.g., sub-1 ms per TTI), and network heterogeneity (different spectrum bands, cell types, and vendor equipment). A critical yet often overlooked barrier lies in the computational and memory limitations of RAN baseband hardware, particularly in legacy 4th Generation (4G) systems, which typically lack on-chip neural accelerators. As a result, only lightweight AI models (under 1 Mb and sub-100~μs inference time) can be effectively deployed, limiting both their performance and applicability. However, achieving strong generalization across diverse network conditions often requires large-scale models with substantial resource demands. To address this trade-off, this paper investigates policy distillation in the context of a reinforcement learning-based link adaptation task. We explore two strategies: single-policy distillation, where a scenario-agnostic teacher model is compressed into one generalized student model; and multi-policy distillation, where multiple scenario-specific teachers are consolidated into a single generalist student. Experimental evaluations in a high-fidelity, 5th Generation (5G)-compliant simulator demonstrate that both strategies produce compact student models that preserve the teachers' generalization capabilities while complying with the computational and memory limitations of existing RAN hardware.","short_abstract":"Adopting artificial intelligence (AI) in radio access networks (RANs) presents several challenges, including limited availability of link-level measurements (e.g., CQI reports), stringent real-time processing constraints (e.g., sub-1 ms per TTI), and network heterogeneity (different spectrum bands, cell types, and vend...","url_abs":"https://arxiv.org/abs/2511.06563","url_pdf":"https://arxiv.org/pdf/2511.06563v2","authors":"[\"Sara Khosravi\",\"Burak Demirel\",\"Linghui Zhou\",\"Javier Rasines\",\"Pablo Soldati\"]","published":"2025-11-09T22:48:10Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}