# Generating Long-term Trajectories Using Deep Hierarchical Networks

arXiv: https://arxiv.org/abs/1706.07138 · HTML: https://mlsanity.com/paper/1706.07138/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Jun 2017.

## Built on by (most cited first, 23 of 23)

- [Deep Multi-Task Learning for Joint Localization, Perception, and Prediction](https://mlsanity.com/paper/2101.06720/) (2021)
- [A Comprehensive Review of Computer Vision in Sports: Open Issues, Future Trends and Research Directions](https://mlsanity.com/paper/2203.02281/) (2022)
- [LightTR: A Lightweight Framework for Federated Trajectory Recovery](https://mlsanity.com/paper/2405.03409/) (2024)
- [Unified Uncertainty-Aware Diffusion for Multi-Agent Trajectory Modeling](https://mlsanity.com/paper/2503.18589/) (2025)
- [baller2vec: A Multi-Entity Transformer For Multi-Agent Spatiotemporal Modeling](https://mlsanity.com/paper/2102.03291/) (2021)
- [Interaction Modeling with Multiplex Attention](https://mlsanity.com/paper/2208.10660/) (2022)
- [Trajectory Prediction for Autonomous Driving: Progress, Limitations, and Future Directions](https://mlsanity.com/paper/2503.03262/) (2025)
- [baller2vec++: A Look-Ahead Multi-Entity Transformer For Modeling Coordinated Agents](https://mlsanity.com/paper/2104.11980/) (2021)
- [TranSPORTmer: A Holistic Approach to Trajectory Understanding in Multi-Agent Sports](https://mlsanity.com/paper/2410.17785/) (2024)
- [Adaptive action supervision in reinforcement learning from real-world multi-agent demonstrations](https://mlsanity.com/paper/2305.13030/) (2023)
- [Micro-Macro Spatial-Temporal Graph-based Encoder-Decoder for Map-Constrained Trajectory Recovery](https://mlsanity.com/paper/2404.19141/) (2024)
- [FootBots: A Transformer-based Architecture for Motion Prediction in Soccer](https://mlsanity.com/paper/2406.19852/) (2024)
- [Data-driven Analysis for Understanding Team Sports Behaviors](https://mlsanity.com/paper/2102.07545/) (2021)
- [Learning interaction rules from multi-animal trajectories via augmented behavioral models](https://mlsanity.com/paper/2107.05326/) (2021)
- [Group Activity Recognition in Basketball Tracking Data -- Neural Embeddings in Team Sports (NETS)](https://mlsanity.com/paper/2209.00451/) (2022)
- [Mathematical models for off-ball scoring prediction in basketball](https://mlsanity.com/paper/2406.08749/) (2024)
- [JointDiff: Bridging Continuous and Discrete in Multi-Agent Trajectory Generation](https://mlsanity.com/paper/2509.22522/) (2025)
- [Heteroscedastic Diffusion for Multi-Agent Trajectory Modeling](https://mlsanity.com/paper/2605.10717/) (2026)
- [About latent roles in forecasting players in team sports](https://mlsanity.com/paper/2304.08272/) (2023)
- [Analyzing Micro-Founded General Equilibrium Models with Many Agents using Deep Reinforcement Learning](https://mlsanity.com/paper/2201.01163/) (2022)
- [Multi-Modal Soccer Scene Analysis with Masked Pre-Training](https://mlsanity.com/paper/2512.19528/) (2025)
- [SportsNGEN: Sustained Generation of Realistic Multi-player Sports Gameplay](https://mlsanity.com/paper/2403.12977/) (2024)
- [CourtMotion: Learning Event-Driven Motion Representations from Skeletal Data for Basketball](https://mlsanity.com/paper/2512.01478/) (2025)

## Abstract (arXiv)

We study the problem of modeling spatiotemporal trajectories over long time horizons using expert demonstrations. For instance, in sports, agents often choose action sequences with long-term goals in mind, such as achieving a certain strategic position. Conventional policy learning approaches, such as those based on Markov decision processes, generally fail at learning cohesive long-term behavior in such high-dimensional state spaces, and are only effective when myopic modeling lead to the desired behavior. The key difficulty is that conventional approaches are "shallow" models that only learn a single state-action policy. We instead propose a hierarchical policy class that automatically reasons about both long-term and short-term goals, which we instantiate as a hierarchical neural network. We showcase our approach in a case study on learning to imitate demonstrated basketball trajectories, and show that it generates significantly more realistic trajectories compared to non-hierarchical baselines as judged by professional sports analysts.