[Computational Learning Theory Team Seminar] Talks of Dr. Dimitra Tsigkari and Dr. Fion Mc Inerney (Telefonica, Spain)
Two Telefonica researchers visit RIKEN AIP to talk Split Learning network optimization and non-clashing teaching theory in graphs.
- When
- Fri, May 22, 2026 · 13:30–15:30 JST
- Where
- Hybrid
- Organizer
- RIKEN Center for Advanced Intelligence Project
- Language
- EN
- Source
- Doorkeeper
Summary
RIKEN AIP's Computational Learning Theory Team hosts two research talks from Telefonica researchers visiting from Spain. Dr. Dimitra Tsigkari presents "Network Optimization for Split Learning," covering optimization challenges in Split Learning frameworks where heterogeneous clients offload part of model training to a server, with results drawn from INFOCOM 2024/2026 and AAAI 2026 papers (including work on catastrophic forgetting under data heterogeneity).
Dr. Fion Mc Inerney follows with "Non-Clashing Teaching in Graphs," a talk on batch machine teaching theory. The work studies the teaching dimension for balls in graphs, presenting NP-hardness, W[1]-hardness, FPT algorithms, and combinatorial results for trees, cycles, and interval graphs, drawing from COLT 2024, ICLR 2025, and ICLR 2026 papers.
Each talk runs 45 minutes including Q&A. The seminar is delivered via Zoom for all registered participants, with on-site attendance at the RIKEN Nihonbashi Office Open Space restricted to RIKEN members.
About the community
RIKEN AIP's Computational Learning Theory Team runs research seminars hosting invited speakers from international labs. Sessions are technical, paper-driven, and aimed at machine learning researchers and graduate students interested in learning theory, optimization, and theoretical foundations of AI.
#machine-learning#learning-theory#split-learning#network-optimization#graph-theory#research-seminar#riken