# Online Residual Learning from Offline Experts for Pedestrian Tracking

arXiv: https://arxiv.org/abs/2409.04069 · HTML: https://mlsanity.com/paper/2409.04069/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 50 references that are in the map.
- First posted on arXiv 6 Sep 2024.

## Builds on (most cited first, 19 of 19)

- [AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting](https://mlsanity.com/paper/2103.14023/) (2021)
- [Multimodal Trajectory Predictions for Autonomous Driving using Deep Convolutional Networks](https://mlsanity.com/paper/1809.10732/) (2018)
- [Scaling Laws for Fine-Grained Mixture of Experts](https://mlsanity.com/paper/2402.07871/) (2024)
- [From Goals, Waypoints & Paths To Long Term Human Trajectory Forecasting](https://mlsanity.com/paper/2012.01526/) (2020)
- [Hybrid RL: Using Both Offline and Online Data Can Make RL Efficient](https://mlsanity.com/paper/2210.06718/) (2022)
- [Online Learning under Delayed Feedback](https://mlsanity.com/paper/1306.0686/) (2013)
- [Cautious Model Predictive Control using Gaussian Process Regression](https://mlsanity.com/paper/1705.10702/) (2017)
- [Leveraging Offline Data in Online Reinforcement Learning](https://mlsanity.com/paper/2211.04974/) (2022)
- [Logarithmic Regret for Adversarial Online Control](https://mlsanity.com/paper/2003.00189/) (2020)
- [PSI: A Benchmark for Human Interpretation and Response in Traffic Interactions](https://mlsanity.com/paper/2112.02604/) (2021)
- [Online Gradient Boosting](https://mlsanity.com/paper/1506.04820/) (2015)
- [Trading-Off Static and Dynamic Regret in Online Least-Squares and Beyond](https://mlsanity.com/paper/1909.03118/) (2019)
- [Learn Fast, Forget Slow: Safe Predictive Learning Control for Systems with Unknown and Changing Dynamics Performing Repetitive Tasks](https://mlsanity.com/paper/1810.06681/) (2018)
- [Analyzing Human Models that Adapt Online](https://mlsanity.com/paper/2103.05746/) (2021)
- [Boosting for Control of Dynamical Systems](https://mlsanity.com/paper/1906.08720/) (2019)
- [Predictive Linear Online Tracking for Unknown Targets](https://mlsanity.com/paper/2402.10036/) (2024)
- [Learning-Based Repetitive Precision Motion Control with Mismatch Compensation](https://mlsanity.com/paper/2111.10246/) (2021)
- [EANet: Expert Attention Network for Online Trajectory Prediction](https://mlsanity.com/paper/2309.05683/) (2023)
- [Fusing Multiple Algorithms for Heterogeneous Online Learning](https://mlsanity.com/paper/2312.05432/) (2023)

## Built on by (most cited first, 1 of 1)

- [Adaptive prediction theory combining offline and online learning](https://mlsanity.com/paper/2512.00342/) (2025)

## Abstract (arXiv)

In this paper, we consider the problem of predicting unknown targets from data. We propose Online Residual Learning (ORL), a method that combines online adaptation with offline-trained predictions. At a lower level, we employ multiple offline predictions generated before or at the beginning of the prediction horizon. We augment every offline prediction by learning their respective residual error concerning the true target state online, using the recursive least squares algorithm. At a higher level, we treat the augmented lower-level predictors as experts, adopting the Prediction with Expert Advice framework. We utilize an adaptive softmax weighting scheme to form an aggregate prediction and provide guarantees for ORL in terms of regret. We employ ORL to boost performance in the setting of online pedestrian trajectory prediction. Based on data from the Stanford Drone Dataset, we show that ORL can demonstrate best-of-both-worlds performance.