Mert Albaba

Mert Albaba, Chenhao Li, Markos Diomataris et al.

Acquiring physically plausible motor skills across diverse and unconventional morphologies-including humanoid robots, quadrupeds, and animals-is essential for advancing character simulation and robotics. Traditional methods, such as reinforcement learning (RL) are task- and body-specific, require extensive reward function engineering, and do not generalize well. Imitation learning offers an alternative but relies heavily on high-quality expert demonstrations, which are difficult to obtain for non-human morphologies. Video diffusion models, on the other hand, are capable of generating realistic videos of various morphologies, from humans to ants. Leveraging this capability, we propose a data-independent approach for skill acquisition that learns 3D motor skills from 2D-generated videos, with generalization capability to unconventional and non-human forms. Specifically, we guide the imitation learning process by leveraging vision transformers for video-based comparisons by calculating pair-wise distance between video embeddings. Along with video-encoding distance, we also use a computed similarity between segmented video frames as a guidance reward. We validate our method on locomotion tasks involving unique body configurations. In humanoid robot locomotion tasks, we demonstrate that 'No-data Imitation Learning' (NIL) outperforms baselines trained on 3D motion-capture data. Our results highlight the potential of leveraging generative video models for physically plausible skill learning with diverse morphologies, effectively replacing data collection with data generation for imitation learning.

Marco Bagatella, Mert Albaba, Jonas Hübotter et al.

Foundation models compress a large amount of information in a single, large neural network, which can then be queried for individual tasks. There are strong parallels between this widespread framework and offline goal-conditioned reinforcement learning algorithms: a universal value function is trained on a large number of goals, and the policy is evaluated on a single goal in each test episode. Extensive research in foundation models has shown that performance can be substantially improved through test-time training, specializing the model to the current goal. We find similarly that test-time offline reinforcement learning on experience related to the test goal can lead to substantially better policies at minimal compute costs. We propose a novel self-supervised data selection criterion, which selects transitions from an offline dataset according to their relevance to the current state and quality with respect to the evaluation goal. We demonstrate across a wide range of high-dimensional loco-navigation and manipulation tasks that fine-tuning a policy on the selected data for a few gradient steps leads to significant performance gains over standard offline pre-training. Our goal-conditioned test-time training (GC-TTT) algorithm applies this routine in a receding-horizon fashion during evaluation, adapting the policy to the current trajectory as it is being rolled out. Finally, we study compute allocation at inference, demonstrating that, at comparable costs, GC-TTT induces performance gains that are not achievable by scaling model size.

Mert Albaba, Sammy Christen, Thomas Langarek et al.

Acquiring complex behaviors is essential for artificially intelligent agents, yet learning these behaviors in high-dimensional settings poses a significant challenge due to the vast search space. Traditional reinforcement learning (RL) requires extensive manual effort for reward function engineering. Inverse reinforcement learning (IRL) uncovers reward functions from expert demonstrations but relies on an iterative process that is often computationally expensive. Imitation learning (IL) provides a more efficient alternative by directly comparing an agent's actions to expert demonstrations; however, in high-dimensional environments, such direct comparisons often offer insufficient feedback for effective learning. We introduce RILe (Reinforced Imitation Learning), a framework that combines the strengths of imitation learning and inverse reinforcement learning to learn a dense reward function efficiently and achieve strong performance in high-dimensional tasks. RILe employs a novel trainer-student framework: the trainer learns an adaptive reward function, and the student uses this reward signal to imitate expert behaviors. By dynamically adjusting its guidance as the student evolves, the trainer provides nuanced feedback across different phases of learning. Our framework produces high-performing policies in high-dimensional tasks where direct imitation fails to replicate complex behaviors. We validate RILe in challenging robotic locomotion tasks, demonstrating that it significantly outperforms existing methods and achieves near-expert performance across multiple settings.

Mert Albaba, Yildiray Yildiz

Predicting the outcomes of cyber-physical systems with multiple human interactions is a challenging problem. This article reviews a game theoretical approach to address this issue, where reinforcement learning is employed to predict the time-extended interaction dynamics. We explain that the most attractive feature of the method is proposing a computationally feasible approach to simultaneously model multiple humans as decision makers, instead of determining the decision dynamics of the intelligent agent of interest and forcing the others to obey certain kinematic and dynamic constraints imposed by the environment. We present two recent exploitations of the method to model 1) unmanned aircraft integration into the National Airspace System and 2) highway traffic. We conclude the article by providing ongoing and future work about employing, improving and validating the method. We also provide related open problems and research opportunities.