Optimizing Materials Discovery with Quantum and Classical Machine Learning Techniques

BASF is the largest chemical producer in the world, constantly innovating and developing new materials for a wide range of sectors. These sectors include consumer goods, transportation, healthcare, agriculture, and energy among others. In their pursuit of sustainable and innovative new materials, BASF is always looking for ways to leverage advanced technologies such as AI and quantum techniques. Their goal is to enhance their existing cheminformatics solutions, particularly machine learning models that can predict the molecular properties of new materials. They aim to make their materials discovery process more efficient and effective, and to optimize their operations across the value chain.

BASF, the world's largest chemical producer, is constantly innovating and developing new materials for various sectors including consumer goods, transportation, healthcare, agriculture, and energy. The challenge lies in their pursuit of sustainable and innovative new materials. BASF is keen on exploring how AI and quantum techniques can be utilized on today's classical computers to enhance existing cheminformatics solutions. Specifically, they are interested in machine learning models that can predict the molecular properties of new materials. The goal is to leverage these advanced technologies to boost their materials discovery process and make it more efficient and effective.

To address this challenge, BASF has partnered with Zapata to benchmark their proprietary quantum-enhanced machine learning techniques against state-of-the-art classical approaches using the Orquestra platform. The focus is on exploring machine learning approaches for predicting molecular properties, with feature selection and classification as a sub-routine of a supervised learning algorithm. This collaboration aims to enhance the materials discovery process by leveraging advanced machine learning techniques. In addition to this, BASF is also exploring how generative AI methods like Zapata’s GEO can help optimize their operations across the value chain. This includes everything from raw material sourcing to the location of production facilities and beyond. Using Zapata’s generator-enhanced optimization (GEO) technique, they can train generative models on the best available solutions to these problems and generate new, previously unconsidered solutions.