W3: Machine Learning for Wireless Communications
Time: 9:00–12:30
Room: Kaimuki 2
Abstract: During the last decade, tremendous research efforts have been put into machine learning (ML) which led to breakthroughs in many fields such as computer vision, natural language processing, pattern recognition, and game play. Recently, there has been a growing interest in ML also for communication systems. Motivated by encouraging preliminary results and the game changing progress observed in other domains, it is believed that ML could lead to significant advances in communication systems with a long-lasting impact. Moreover, ML enables novel data-driven approaches to system design which render accurate models – when available – less relevant.
Despite the recent vivid interest for ML in communications, the full potential and limitations of ML in this domain have not been yet fully understood. Therefore, we solicit novel contributions on the topics listed below.
Topics
- Deep learning for channel coding and transceiver design
- Neural networks for wireless communication (end-to-end learning, autoencoders, generative adversarial networks, etc.)
- Information bottleneck methods for communications
- Deep learning for radio resource management
- Deep Learning for user behavior and demand prediction
- Deep Learning for user localization and trajectory prediction
- Novel use-case leveraging machine learning
- Deep learning for MIMO systems
- Deep learning for access control
- Deep learning for channel estimation
- Deep learning for URLLC
- On-device machine learning
- Co-design of hardware and machine learning algorithms
- Implementation and acceleration of machine learning algorithms
- Experimental results on applications of AI in wireless systems
- Security in ML-based communication systems
- New data sets and ML challenges in wireless systems
- ML-augmented algorithms and models
Workshop home: https://vtc19wkshpmlwic.wixsite.com/wkshpmlwic
Submit a paper: https://vtc2019f-rr-wks.trackchair.com/track/1794