New AI Tool by Coursera Aims to Detect Plagiarism Without Bias, Pleasing Tutors Everywhere!
June 14, 2024Tired of Getting Lost? Tern AI Unveils GPS Alternative – Say Hello to Safer, Smarter Navigation!
June 17, 2024AI Assistant Revolutionizes Organic Chemistry
Researchers at the University of Liverpool, in collaboration with IBM, have developed an AI assistant to help synthetic chemists discover new, cost-effective ways to create organic molecules (Phys.org). The AI uses Bayesian optimization, a computational method adept at handling complex problems with numerous variables. This approach has already led to the discovery of a catalyst for a carbon-carbon bond-forming reaction that avoids using the expensive metal iridium, which costs around $170 per gram.
Smarter Experimentation Through Bayesian Optimization
Bayesian optimization excels in processing large, interconnected variables, making it highly suitable for chemistry research. By incorporating known chemistry rules, such as acid-base neutralisation, the AI can suggest more effective experiments. For instance, it evaluates physical properties like light absorption to infer the efficacy of chemical combinations, thus avoiding random permutations or overly simplistic rules. This smarter method allows chemists to integrate their insights with the algorithm’s suggestions, enhancing the experimental process.
Significant Advances in Catalyst Formulation
In practice, this AI tool guided the team to develop highly effective iridium-free catalyst formulations through 107 experiments, a significant reduction from the 4500 possible combinations. This method is potentially transferable to other chemical reactions where computing relevant physical properties can guide optimization. Professor Andrew Cooper emphasised that the tool serves as an AI assistant, not a replacement for human intelligence, highlighting its role in augmenting rather than replacing human chemists’ expertise.
(Visit Phys.org for the full story)
*An AI tool was used to add an extra layer to the editing process for this story.