The field of control and learning is witnessing significant advancements, with a focus on developing innovative methods for implicit communication, iterative learning control, and imitation learning. Researchers are exploring new approaches to enable efficient communication and control in complex systems, such as linear quadratic Gaussian control systems. Meanwhile, iterative learning control methods are being developed to address the challenges of unknown dynamics and manual parameter tuning in multiple-input multiple-output systems. Imitation learning is also gaining attention, with a focus on learning from observation and self-evolved imitation learning. Noteworthy papers in this area include:

• Implicit Communication in Linear Quadratic Gaussian Control Systems, which formalizes the notion of implicit channel capacity and characterizes it in various settings.
• Dual Iterative Learning Control for Multiple-Input Multiple-Output Dynamics with Validation in Robotic Systems, which presents a novel data-driven iterative learning scheme for simultaneous tracking control and model learning.
• Self-evolved Imitation Learning in Simulated World, which proposes a framework for progressively improving a few-shot model through simulator interactions.
• RoboSSM: Scalable In-context Imitation Learning via State-Space Models, which introduces a scalable recipe for in-context imitation learning based on state-space models.