arXiv subjects

Alexander Zap

Publications and source records attributed to Alexander Zap.

4 recordsLinked to original sources

arXiv2026

Multi-Task Learning for Heterogeneous Prediction from Video Game State with Transfer Learning

Multi-task learning (MTL) is a promising approach for prediction tasks derived from video game state data, as modern game telemetry provides multiple related supervision signals from the same structured observations. We study whether a shared model trained jointly across tasks in team-based multiplayer games can improve generalization while reducing training and inference cost compared to specialized single-task models. We adapt a multimodal architecture for endpoint prediction to a general multi-task setting that combines rasterized vision inputs, global match context, and per-unit state information through an image encoder and attention-based interaction modeling. Experiments on a large proprietary World of Tanks dataset compare single-task and multi-task training, evaluate weighting strategies for mixed losses and conflicting gradients, and test pre-training/fine-tuning under limited target-data regimes. We also examine within-game transfer across game maps under structured environment shift.

cs.LG↗

arXiv2025

A Multimodal Architecture for Endpoint Position Prediction in Team-based Multiplayer Games

Understanding and predicting player movement in multiplayer games is crucial for achieving use cases such as player-mimicking bot navigation, preemptive bot control, strategy recommendation, and real-time player behavior analytics. However, the complex environments allow for a high degree of navigational freedom, and the interactions and team-play between players require models that make effective use of the available heterogeneous input data. This paper presents a multimodal architecture for predicting future player locations on a dynamic time horizon, using a U-Net-based approach for calculating endpoint location probability heatmaps, conditioned using a multimodal feature encoder. The application of a multi-head attention mechanism for different groups of features allows for communication between agents. In doing so, the architecture makes efficient use of the multimodal game state including image inputs, numerical and categorical features, as well as dynamic game data. Consequently, the presented technique lays the foundation for various downstream tasks that rely on future player positions such as the creation of player-predictive bot behavior or player anomaly detection.

cs.CV↗

arXiv2025

Where are the Frontlines? A Visualization Approach for Map Control in Team-Based Games

A central area of interest in many competitive online games is spatial behavior which due to its complexity can be difficult to visualize. Such behaviors of interest include not only overall movement patterns but also being able to understand which player or team is exerting control over an area to inform decision-making. Map control can, however, be challenging to quantify. In this paper, we propose a method for calculating frontlines and first efforts towards a visualization of them. The visualization can show map control and frontlines at a specific time point or changes of these over time. For this purpose, it utilizes support vector machines to derive frontlines from unit positions. We illustrate our algorithm and visualization with examples based on the team-based online game World of Tanks.

cs.HC↗

arXiv2019

Deep Ordinal Reinforcement Learning

Reinforcement learning usually makes use of numerical rewards, which have nice properties but also come with drawbacks and difficulties. Using rewards on an ordinal scale (ordinal rewards) is an alternative to numerical rewards that has received more attention in recent years. In this paper, a general approach to adapting reinforcement learning problems to the use of ordinal rewards is presented and motivated. We show how to convert common reinforcement learning algorithms to an ordinal variation by the example of Q-learning and introduce Ordinal Deep Q-Networks, which adapt deep reinforcement learning to ordinal rewards. Additionally, we run evaluations on problems provided by the OpenAI Gym framework, showing that our ordinal variants exhibit a performance that is comparable to the numerical variations for a number of problems. We also give first evidence that our ordinal variant is able to produce better results for problems with less engineered and simpler-to-design reward signals.

cs.LG↗