Artificial Intelligence

The main focus of the Artificial Intelligence department are the theoretical and algorithmic foundations of machine learning, the development of novel methods and models of deep learning and the application of these AI techniques in practice. A core topic is the explainability and interpretability of deep learning models. With LRP and SpRAy, the department has developed pioneering algorithms for the explanation of deep neural networks, which have already been used successfully in various scientific and industrial applications. The development of methods for compression of neural networks and for efficient federated learning are two further central research topics of the department. With DeepCABAC, new trellis-based quantization techniques and various other contributions, the department, in close collaboration with the department of "Video Communication and Applications", has made a significant contribution to shaping the MPEG NNC standard. The robustness and reliability of machine learning algorithms, the investigation of quality criteria for the certification of AI procedures and the adaptation of these models for practical applications, especially in medicine, are also among the core interests of the department. As part of BIFOLD and the ELLIS Unit Berlin, the department is deeply anchored in Berlin's AI research landscape.