Bio-inspired (neuromorphic) methodologies, machine learning algorithms and new computational concepts are emerging as novel scientific tools and technology paradigms. Blending machine learning with photonics enables the development of new research and engineering techniques that benefit both from the speed and parallelism inherent to optical systems and the ability of machine learning to infer from data and automatically improve system performance. Nonlinear photonics often features complex (e.g., spatio-temporal in multimode systems) light dynamics, and deals with systems that cannot be easily comprehended and controlled. Therefore, applications of the machine learning in photonics is plausible due to its capability to deal with complex nonlinear systems, whilst staying flexible and re-adaptable. Additionally, photonic devices and systems operating at high speed can quickly generate a vast amount of data. This makes them well-suited to the application of various data-based machine learning algorithms that improve performance with increasing available data sets. Therefore, combination of photonics and machine learning can naturally bring about new ideas, theories and devices, as well as novel concepts for understanding and description of light-related phenomena.

The summer school will bring together world-leading experts in nonlinear photonics, machine learning, neuromorphic photonics and optical computational concepts.