complementary ai systems

Future robots need to be robust and adaptable, and new design approaches are needed for new production methods. I will talk about my research in using evolutionary algorithms and biologically inspired methods with the aim of having more intelligent, robust, and adaptive behavior in robots. I will give a short introduction to some of the algorithms and show how we, at the University of Oslo, apply them in our research platforms for exploring automatic robot design and adaptation. Here, we take an embodied AI approach and aim to co-design the body and the behavior of the robots, such that they are well fit for their intended environments and tasks. These approaches are still at a fundamental research stage, and I will discuss potential future application areas, as well as challenges and opportunities related to the sustainability of these AI systems.

Learning objectives:

By the end of the session, participants will be able to:

• List some bio-inspired AI approaches for robot design and adaptation.
• Discuss challenges and opportunities related to algorithmic robot design.

Recommended mastery level:

• Some undergraduate level engineering or computer science skills would be beneficial, although not absolutely necessary.

Artificial Intelligence is transforming how we approach chemical research and synthesis. By teaching language models to understand and generate the language of chemistry, we have developed complementary AI systems that bridge the gap between computational design and experimental reality.

Our large language model system, ChemCrow, represents one of the first demonstrations of an AI system directly controlling robotic synthesis platforms, successfully executing the synthesis of compounds including organocatalysts and chromophores.

Complementing this, our small language model system, Saturn, currently the most sample-efficient molecular design algorithm, enables precise molecular generation with built-in synthesizability constraints. Saturn’s innovations include direct optimization against retrosynthetic predictions and integration of building block availability, ensuring that generated molecules are practically accessible.

Our work demonstrates how different scales of language models can work together to transform chemical research, from initial molecular design through to physical synthesis, potentially revolutionizing drug discovery, catalysis, and materials development.