Patrick Lucey
YOU?
Author Swipe
View article: Event2Tracking: Reconstructing Multi-Agent Soccer Trajectories Using Long-Term Multimodal Context
Event2Tracking: Reconstructing Multi-Agent Soccer Trajectories Using Long-Term Multimodal Context Open
Soccer is a rich testbed for studying multi-agent adversarial systems. In this work we focus on the task of reconstructing the noisy trajectories of soccer agents (players and the ball). Previous works that model the behaviours of agents i…
View article: Characterizing Multi-Agent Team Behavior from Partial Team Tracings: Evidence from the English Premier League
Characterizing Multi-Agent Team Behavior from Partial Team Tracings: Evidence from the English Premier League Open
Real-world AI systems have been recently deployed which can automatically analyze the plan and tactics of tennis players. As the game-state is updated regularly at short intervals (i.e. point-level), a library of successful and unsuccessfu…
View article: You Cannot Do That Ben Stokes: Dynamically Predicting Shot Type in Cricket Using a Personalized Deep Neural Network
You Cannot Do That Ben Stokes: Dynamically Predicting Shot Type in Cricket Using a Personalized Deep Neural Network Open
The ability to predict what shot a batsman will attempt given the type of ball and match situation is both one of the most challenging and strategically important tasks in cricket. The goal of the batsman is to score as many runs without b…
View article: Improved Structural Discovery and Representation Learning of Multi-Agent Data
Improved Structural Discovery and Representation Learning of Multi-Agent Data Open
Central to all machine learning algorithms is data representation. For multi-agent systems, selecting a representation which adequately captures the interactions among agents is challenging due to the latent group structure which tends to …
View article: Rugby-Bot: Utilizing Multi-Task Learning & Fine-Grained Features for Rugby League Analysis
Rugby-Bot: Utilizing Multi-Task Learning & Fine-Grained Features for Rugby League Analysis Open
Sporting events are extremely complex and require a multitude of metrics to accurate describe the event. When making multiple predictions, one should make them from a single source to keep consistency across the predictions. We present a m…
View article: Rugby-Bot: Utilizing Multi-Task Learning & Fine-Grained Features for Rugby League Analysis.
Rugby-Bot: Utilizing Multi-Task Learning & Fine-Grained Features for Rugby League Analysis. Open
Sporting events are extremely complex and require a multitude of metrics to accurate describe the event. When making multiple predictions, one should make them from a single source to keep consistency across the predictions. We present a m…
View article: Generating Multi-Agent Trajectories using Programmatic Weak Supervision
Generating Multi-Agent Trajectories using Programmatic Weak Supervision Open
We study the problem of training sequential generative models for capturing coordinated multi-agent trajectory behavior, such as offensive basketball gameplay. When modeling such settings, it is often beneficial to design hierarchical mode…
View article: Generating Multi-Agent Trajectories using Programmatic Weak Supervision
Generating Multi-Agent Trajectories using Programmatic Weak Supervision Open
We study the problem of training sequential generative models for capturing coordinated multi-agent trajectory behavior, such as offensive basketball gameplay. When modeling such settings, it is often beneficial to design hierarchical mode…
View article: Generative Multi-Agent Behavioral Cloning
Generative Multi-Agent Behavioral Cloning Open
We propose and study the problem of generative multi-agent behavioral cloning, where the goal is to learn a generative, i.e., non-deterministic, multi-agent policy from pre-collected demonstration data. Building upon advances in deep gener…
View article: Fine-Grained Retrieval of Sports Plays using Tree-Based Alignment of Trajectories
Fine-Grained Retrieval of Sports Plays using Tree-Based Alignment of Trajectories Open
We propose a novel method for effective retrieval of multi-agent spatiotemporal tracking data. Retrieval of spatiotemporal tracking data offers several unique challenges compared to conventional text-based retrieval settings. Most notably,…
View article: Generating Long-term Trajectories Using Deep Hierarchical Networks
Generating Long-term Trajectories Using Deep Hierarchical Networks Open
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 strat…
View article: Coordinated Multi-Agent Imitation Learning
Coordinated Multi-Agent Imitation Learning Open
We study the problem of imitation learning from demonstrations of multiple coordinating agents. One key challenge in this setting is that learning a good model of coordination can be difficult, since coordination is often implicit in the d…
View article: DATA-DRIVEN GHOSTING USING DEEP IMITATION LEARNING
DATA-DRIVEN GHOSTING USING DEEP IMITATION LEARNING Open
Current state-of-the-art sports metrics such as “Wins-above-Replacement” in baseball, “Expected Point Value” in basketball, and “Expected Goal Value” in soccer and hockey are now commonplace in performance analysis. These measures have enh…
View article: Predicting Shot Making in Basketball Learnt from Adversarial Multiagent\n Trajectories
Predicting Shot Making in Basketball Learnt from Adversarial Multiagent\n Trajectories Open
In this paper, we predict the likelihood of a player making a shot in\nbasketball from multiagent trajectories. Previous approaches to similar\nproblems center on hand-crafting features to capture domain specific knowledge.\nAlthough intui…
View article: Predicting Shot Making in Basketball Learnt from Adversarial Multiagent Trajectories
Predicting Shot Making in Basketball Learnt from Adversarial Multiagent Trajectories Open
In this paper, we predict the likelihood of a player making a shot in basketball from multiagent trajectories. Previous approaches to similar problems center on hand-crafting features to capture domain specific knowledge. Although intuitiv…
View article: Predicting Shot Making in Basketball using Convolutional Neural Networks Learnt from Adversarial Multiagent Trajectories.
Predicting Shot Making in Basketball using Convolutional Neural Networks Learnt from Adversarial Multiagent Trajectories. Open
In this paper, we predict the likelihood of a player making a shot in basketball from multiagent trajectories. Previous approaches to similar problems center on hand-crafting features to capture domain specific knowledge. Although intuitiv…
View article: Softstar: heuristic-guided probabilistic inference
Softstar: heuristic-guided probabilistic inference Open
Recent machine learning methods for sequential behavior prediction estimate the motives of behavior rather than the behavior itself. This higher-level abstraction improves generalization in different prediction settings, but computing pred…
View article: Discovering methods of scoring in soccer using tracking data
Discovering methods of scoring in soccer using tracking data Open
In soccer, when analyzing the performance of a team one of the key events to analyze is that of shots and goal-scoring. With the availability of fine-grained player and ball tracking data, it is now possible to find the common patterns a t…