Graph Conditional Variational Models: Too Complex for Multiagent Trajectories?

Research output: Journal contributionsConference article in journalResearchpeer-review


Recent advances in modeling multiagent trajectories combine graph architectures such as graph neural networks (GNNs) with conditional variational models (CVMs) such as variational RNNs (VRNNs). Originally, CVMs have been proposed to facilitate learning with multi-modal and structured data and thus seem to perfectly match the requirements of multi-modal multiagent trajectories with their structured output spaces. Empirical results of VRNNs on trajectory data support this assumption. In this paper, we revisit experiments and proposed architectures with additional rigour, ablation runs and baselines. In contrast to common belief, we show that prior results with CVMs on trajectory data might be misleading. Given a neural network with a graph architecture and/or structured output function, variational autoencoding does not seem to contribute statistically significantly to empirical performance. Instead, we show that well-known emission functions do contribute, while coming with less complexity, engineering and computation time.
Original languageEnglish
JournalProceedings of Machine Learning Research
Pages (from-to)136-147
Number of pages12
Publication statusPublished - 2020
EventVirtual NeurIPS 2020: Neural Information Processing Systems Online Conference 2020 - digital
Duration: 06.12.202012.12.2020
Conference number: 34

Bibliographical note

Proceedings on "I Can't Believe It's Not Better!" at NeurIPS Workshops, 2020.