Abstract
Learning-based multi-agent communication under limited bandwidth does not only require deciding what to communicate, but also structuring messages so that partial transmissions remain useful. We study this problem under prefix truncation, where only the first part of each message is received. To address it, we propose \textbf{AH-VIB}, an attention-based autoregressive variational communication model that combines a variational information bottleneck (VIB) with sequential message generation and a hierarchical robustness loss. We evaluate AH-VIB on a custom cooperative object-inspection and occupancy-mapping task, where agents equipped with a limited field-of-view sensor coordinate to scan inspection objects in an occupancy-grid world, under variable and fixed bandwidth conditions, and compare it against MADDPG, CommNet, a flat VIB baseline, and an autoregressive MLP ablation. AH-VIB achieves competitive mean return while improving performance reliability under the most constrained bandwidth conditions. These results indicate that AH-VIB improves the reliability and graceful degradation of learned communication under bandwidth constraints.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Vindbjerg, L. K., Zhang, Q., Brodskiy, Y., & Esterle, L. (2026). Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth. https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth
MLA 9
Vindbjerg, Lukas Koch, et al. "Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth." https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth.
Chicago (author–date)
Vindbjerg, Lukas Koch, Qi Zhang, Yury Brodskiy, and Lukas Esterle. 2026. "Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth." https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth.
Harvard
Vindbjerg, L. K., Zhang, Q., Brodskiy, Y. and Esterle, L. (2026) 'Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth', Available at: https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth.
Vancouver
Vindbjerg LK, Zhang Q, Brodskiy Y, Esterle L. Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth. https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth
IEEE
L. K. Vindbjerg, Q. Zhang, Y. Brodskiy, and L. Esterle, "Attention-based Hierarchical Variational Information Bottleneck for Robust Multi-Agent Communication under Variable Bandwidth," https://omanscience.com/en/articles/attention-based-hierarchical-variational-information-bottleneck-for-robust-multi-agent-communication-under-variable-bandwidth.