Game design hinges on understanding how static rules and content translate into dynamic player behavior - something modern generative systems that inspect only a game's code or assets struggle to capture. We present an automated design iteration framework that closes this gap by pairing a reinforcement learning (RL) agent, which playtests the game, with a large multimodal model (LMM), which revises the game based on what the agent does. In each loop the RL player completes several episodes, producing (i) numerical play metrics and/or (ii) a compact image strip summarising recent video frames. The LMM designer receives a gameplay goal and the current game configuration, analyses the play traces, and edits the configuration to steer future behaviour toward the goal. We demonstrate results that LMMs can reason over behavioral traces supplied by RL agents to iteratively refine game mechanics, pointing toward practical, scalable tools for AI-assisted game design.

Developing agents capable of fluid gameplay in first/third-person games without API access remains a critical challenge in Artificial General Intelligence (AGI). Recent efforts leverage Vision Language Models (VLMs) as direct controllers, frequently pausing the game to analyze screens and plan action through language reasoning. However, this inefficient paradigm fundamentally restricts agents to basic and non-fluent interactions: relying on isolated VLM reasoning for each action makes it impossible to handle tasks requiring high reactivity (e.g., FPS shooting) or dynamic adaptability (e.g., ACT combat). To handle this, we propose a paradigm shift in gameplay agent design: instead of directly controlling gameplay, VLM develops specialized execution modules tailored for tasks like shooting and combat. These modules handle real-time game interactions, elevating VLM to a high-level developer. Building upon this paradigm, we introduce GameSense, a gameplay agent framework where VLM develops task-specific game sense modules by observing task execution and leveraging vision tools and neural network training pipelines. These modules encapsulate action-feedback logic, ranging from direct action rules to neural network-based decisions. Experiments demonstrate that our framework is the first to achieve fluent gameplay in diverse genres, including ACT, FPS, and Flappy Bird, setting a new benchmark for game-playing agents.

Reinforcement learning is a powerful approach for training an optimal policy to solve complex problems in a given system. This project aims to demonstrate the application of reinforcement learning in stochastic process environments with missing information, using Flappy Bird and a newly developed stock trading environment as case studies. We evaluate various structures of Deep Q-learning networks and identify the most suitable variant for the stochastic process environment. Additionally, we discuss the current challenges and propose potential improvements for further work in environment-building and reinforcement learning techniques.

For over a decade now, robotics and the use of artificial agents have become a common thing.Testing the performance of new path finding or search space optimization algorithms has also become a challenge as they require simulation or an environment to test them.The creation of artificial environments with artificial agents is one of the methods employed to test such algorithms.Games have also become an environment to test them.The performance of the algorithms can be compared by using artificial agents that will behave according to the algorithm in the environment they are put in.The performance parameters can be, how quickly the agent is able to differentiate between rewarding actions and hostile actions.This can be tested by placing the agent in an environment with different types of hurdles and the goal of the agent is to reach the farthest by taking decisions on actions that will lead to avoiding all the obstacles.The environment chosen is a game called "Flappy Bird".The goal of the game is to make the bird fly through a set of pipes of random heights.The bird must go in between these pipes and must not hit the top, the bottom, or the pipes themselves.The actions that the bird can take are either to flap its wings or drop down with gravity.The algorithms that are enforced on the artificial agents are NeuroEvolution of Augmenting Topologies (NEAT) and Reinforcement Learning.The NEAT algorithm takes an "N" initial population of artificial agents.They follow genetic algorithms by considering an objective function, crossover, mutation, and augmenting topologies.Reinforcement learning, on the other hand, remembers the state, the action taken at that state, and the reward received for the action taken using a single agent and a Deep Q-learning Network.The performance of the NEAT algorithm improves as the initial population of the artificial agents is increased.

Recent findings suggest that humans deploy cognitive mechanism of physics simulation engines to simulate the physics of objects. We propose a framework for bots to deploy probabilistic programming tools for interacting with intuitive physics environments. The framework employs a physics simulation in a probabilistic way to infer about moves performed by an agent in a setting governed by Newtonian laws of motion. However, methods of probabilistic programs can be slow in such setting due to their need to generate many samples. We complement the model with a model-free approach to aid the sampling procedures in becoming more efficient through learning from experience during game playing. We present an approach where combining model-free approaches (a convolutional neural network in our model) and model-based approaches (probabilistic physics simulation) is able to achieve what neither could alone. This way the model outperforms an all model-free or all model-based approach. We discuss a case study showing empirical results of the performance of the model on the game of Flappy Bird.

The science of applied artificial intelligence doesn't get the same kinds of headlines as the pure research efforts of Google or Facebook or others. Mostly that's because what gets built by companies is obfuscated by those same companies, either for proprietary reasons or because the companies actually have nothing much to speak of. Last week, Babak Hodjat, who runs the machine learning operations of software giant Cognizant Technology Solutions, had something to show, so ZDNet traveled to the loft office near San Francisco's Embarcadero where Hodjat and a team of 18 staffers develop algorithms. The ostensible event was the publication, on the arXiv pre-print server, of a paper showing how Hodjat's style of machine learning could compete with the kind made famous by DeepMind's AlphaZero. Before digging into the paper, ZDNet accepted a challenge against the machine, a game of Flappy Bird.

Deep reinforcement learning has been successfully used in many dynamic decision making domains, especially those with very large state spaces. However, it is also well-known that deep reinforcement learning can be very slow and resource intensive. The resulting system is often brittle and difficult to explain. In this paper, we attempt to address some of these problems by proposing a framework of Rule-interposing Learning (RIL) that embeds high level rules into the deep reinforcement learning. With some good rules, this framework not only can accelerate the learning process, but also keep it away from catastrophic explorations, thus making the system relatively stable even during the very early stage of training. Moreover, given the rules are high level and easy to interpret, they can be easily maintained, updated and shared with other similar tasks.

Self-supervised methods, wherein an agent learns representations solely by observing the results of its actions, become crucial in environments which do not provide a dense reward signal or have labels. In most cases, such methods are used for pretraining or auxiliary tasks for "downstream" tasks, such as control, exploration, or imitation learning. However, it is not clear which method's representations best capture meaningful features of the environment, and which are best suited for which types of environments. We present a small-scale study of self-supervised methods on two visual environments: Flappy Bird and Sonic The Hedgehog. In particular, we quantitatively evaluate the representations learned from these tasks in two contexts: a) the extent to which the representations capture true state information of the agent and b) how generalizable these representations are to novel situations, like new levels and textures. Lastly, we evaluate these self-supervised features by visualizing which parts of the environment they focus on. Our results show that the utility of the representations is highly dependent on the visuals and dynamics of the environment.