Generative Causal Representation Learning for Out-of-Distribution Motion Forecasting

Shayan Shirahmad Gale Bagi, Zahra Gharaee, Oliver Schulte, Mark Crowley
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:31596-31612, 2023.

Abstract

Conventional supervised learning methods typically assume i.i.d samples and are found to be sensitive to out-of-distribution (OOD) data. We propose Generative Causal Representation Learning (GCRL) which leverages causality to facilitate knowledge transfer under distribution shifts. While we evaluate the effectiveness of our proposed method in human trajectory prediction models, GCRL can be applied to other domains as well. First, we propose a novel causal model that explains the generative factors in motion forecasting datasets using features that are common across all environments and with features that are specific to each environment. Selection variables are used to determine which parts of the model can be directly transferred to a new environment without fine-tuning. Second, we propose an end-to-end variational learning paradigm to learn the causal mechanisms that generate observations from features. GCRL is supported by strong theoretical results that imply identifiability of the causal model under certain assumptions. Experimental results on synthetic and real-world motion forecasting datasets show the robustness and effectiveness of our proposed method for knowledge transfer under zero-shot and low-shot settings by substantially outperforming the prior motion forecasting models on out-of-distribution prediction.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-shirahmad-gale-bagi23a, title = {Generative Causal Representation Learning for Out-of-Distribution Motion Forecasting}, author = {Shirahmad Gale Bagi, Shayan and Gharaee, Zahra and Schulte, Oliver and Crowley, Mark}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {31596--31612}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/shirahmad-gale-bagi23a/shirahmad-gale-bagi23a.pdf}, url = {https://proceedings.mlr.press/v202/shirahmad-gale-bagi23a.html}, abstract = {Conventional supervised learning methods typically assume i.i.d samples and are found to be sensitive to out-of-distribution (OOD) data. We propose Generative Causal Representation Learning (GCRL) which leverages causality to facilitate knowledge transfer under distribution shifts. While we evaluate the effectiveness of our proposed method in human trajectory prediction models, GCRL can be applied to other domains as well. First, we propose a novel causal model that explains the generative factors in motion forecasting datasets using features that are common across all environments and with features that are specific to each environment. Selection variables are used to determine which parts of the model can be directly transferred to a new environment without fine-tuning. Second, we propose an end-to-end variational learning paradigm to learn the causal mechanisms that generate observations from features. GCRL is supported by strong theoretical results that imply identifiability of the causal model under certain assumptions. Experimental results on synthetic and real-world motion forecasting datasets show the robustness and effectiveness of our proposed method for knowledge transfer under zero-shot and low-shot settings by substantially outperforming the prior motion forecasting models on out-of-distribution prediction.} }
Endnote
%0 Conference Paper %T Generative Causal Representation Learning for Out-of-Distribution Motion Forecasting %A Shayan Shirahmad Gale Bagi %A Zahra Gharaee %A Oliver Schulte %A Mark Crowley %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-shirahmad-gale-bagi23a %I PMLR %P 31596--31612 %U https://proceedings.mlr.press/v202/shirahmad-gale-bagi23a.html %V 202 %X Conventional supervised learning methods typically assume i.i.d samples and are found to be sensitive to out-of-distribution (OOD) data. We propose Generative Causal Representation Learning (GCRL) which leverages causality to facilitate knowledge transfer under distribution shifts. While we evaluate the effectiveness of our proposed method in human trajectory prediction models, GCRL can be applied to other domains as well. First, we propose a novel causal model that explains the generative factors in motion forecasting datasets using features that are common across all environments and with features that are specific to each environment. Selection variables are used to determine which parts of the model can be directly transferred to a new environment without fine-tuning. Second, we propose an end-to-end variational learning paradigm to learn the causal mechanisms that generate observations from features. GCRL is supported by strong theoretical results that imply identifiability of the causal model under certain assumptions. Experimental results on synthetic and real-world motion forecasting datasets show the robustness and effectiveness of our proposed method for knowledge transfer under zero-shot and low-shot settings by substantially outperforming the prior motion forecasting models on out-of-distribution prediction.
APA
Shirahmad Gale Bagi, S., Gharaee, Z., Schulte, O. & Crowley, M.. (2023). Generative Causal Representation Learning for Out-of-Distribution Motion Forecasting. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:31596-31612 Available from https://proceedings.mlr.press/v202/shirahmad-gale-bagi23a.html.

Related Material