Prof. Krzysztof Krawiec
Poznan University of Technology, PolandTowards Algorithmic Intelligence: Lifting Deep Learning with Neurosymbolic and Physics-Aware Methods
Abstract:
While contemporary deep learning excels at high-dimensional pattern recognition, it remains largely constrained by a reliance on statistical correlations that often fail to capture the underlying causal or algorithmic structures of complex domains. As a result, the models are data-hungry, often struggling with systematic generalization, logical consistency, and adherence to the fundamental laws governing the physical world. This talk delineates a roadmap toward Algorithmic Intelligence, a paradigm that "lifts" standard connectionist architectures by integrating them with symbolic reasoning and the rigorous constraints of physical laws. We examine the frontier of neurosymbolic synthesis, which involves symbolic and programmatic priors into neural architectures, leading to models that don't just predict, but reason. We also explore the embedding of inductive biases derived from physical laws directly into the network's manifold, in particular the use of differentiable renderers and transparent disentanglement of latent representations to enforce physical consistency. We show that such approaches effectively mitigate the optimization plateaus and data inefficiencies prevalent in black-box models in use cases concerning abstract reasoning, interpretation of medical imaging, and remote sensing.
Biography:
Krzysztof Krawiec is a Professor of Computer Science at the Poznan University of Technology in Poland, where he currently serves as the head of the Neurosymbolic Systems Group. His primary research areas include neurosymbolic systems, program synthesis, evolutionary computation, and medical imaging. He has authored over 180 publications on these topics and has received the Fulbright Senior Advanced Research Award, two ACM SIGEVO Impact Awards, and was a visiting professor at the University of California and Massachusetts Institute of Technology. He also served as the general chair of GECCO'21, the largest scientific event in the field of evolutionary computation and as an advisor at the Confederation of Laboratories for Artificial Intelligence in Europe. Krzysztof is also a co-founder of the Center for Artificial Intelligence and Machine learning, part of the Horizon 2020 Foundations of Trustworthy AI project, and an associate editor of Genetic Programming and Evolvable Machines and ACM Transactions on Evolutionary Learning and Optimization. In addition to his academic contributions, Professor Krawiec serves as the Chief AI Officer at Optopol Technology and CTO of Hylomorph Solutions Ltd.
Back to Keynote Speakers