Deep reinforcement learning has successfully been applied for molecular discovery as shown by the Molecule Deep Q-network (MolDQN) algorithm. This algorithm has challenges when applied to optimizing new molecules: training such a model is limited in terms of scalability to larger datasets and the trained model cannot be generalized to different molecules in the same dataset. In this paper, a distributed reinforcement learning algorithm for antioxidants, called DA-MolDQN is proposed to address these problems. State-of-the-art bond dissociation energy (BDE) and ionization potential (IP) predictors are integrated into DA-MolDQN, which are critical chemical properties while optimizing antioxidants. Training time is reduced by algorithmic improvements for molecular modifications. The algorithm is distributed, scalable for up to 512 molecules, and generalizes the model to a diverse set of molecules. The proposed models are trained with a proprietary antioxidant dataset. The results have been reproduced with both proprietary and public datasets. The proposed molecules have been validated with DFT simulations and a subset of them confirmed in public "unseen" datasets. In summary, DA-MolDQN is up to 100x faster than previous algorithms and can discover new optimized molecules from proprietary and public antioxidants.

This article was originally published by The Conversation. Have you ever bitten into a nut or a piece of chocolate expecting a smooth, rich taste only to encounter an unexpected and unpleasant chalky or sour flavor? That taste is rancidity in action, and it affects pretty much every product in your pantry. Now artificial intelligence can help scientists tackle this issue more precisely and efficiently. We're a group of chemists who study ways to extend the life of food products, including those that go rancid.

A fundamental question in natural language processing is - what kind of language structure and semantics is the language model capturing? Graph formats such as knowledge graphs are easy to evaluate as they explicitly express language semantics and structure. This study evaluates the semantics encoded in the self-attention transformers by leveraging explicit knowledge graph structures. We propose novel metrics to measure the reconstruction error when providing graph path sequences from a knowledge graph and trying to reproduce/reconstruct the same from the outputs of the self-attention transformer models. The opacity of language models has an immense bearing on societal issues of trust and explainable decision outcomes. Our findings suggest that language models are models of stochastic control processes for plausible language pattern generation. However, they do not ascribe object and concept-level meaning and semantics to the learned stochastic patterns such as those described in knowledge graphs. Furthermore, to enable robust evaluation of concept understanding by language models, we construct and make public an augmented language understanding benchmark built on the General Language Understanding Evaluation (GLUE) benchmark. This has significant application-level user trust implications as stochastic patterns without a strong sense of meaning cannot be trusted in high-stakes applications.

Quantum and quantum-inspired machine learning has emerged as a promising and challenging research field due to the increased popularity of quantum computing, especially with near-term devices. Theoretical contributions point toward generative modeling as a promising direction to realize the first examples of real-world quantum advantages from these technologies. A few empirical studies also demonstrate such potential, especially when considering quantum-inspired models based on tensor networks. In this work, we apply tensor-network-based generative models to the problem of molecular discovery. In our approach, we utilize two small molecular datasets: a subset of $4989$ molecules from the QM9 dataset and a small in-house dataset of $516$ validated antioxidants from TotalEnergies. We compare several tensor network models against a generative adversarial network using different sample-based metrics, which reflect their learning performances on each task, and multiobjective performances using $3$ relevant molecular metrics per task. We also combined the output of the models and demonstrate empirically that such a combination can be beneficial, advocating for the unification of classical and quantum(-inspired) generative learning.

TORONTO: Scientists have combined radar and artificial intelligence (AI) technologies to detect changes in glucose levels, an advance that may help diabetics monitor their blood sugar without painful finger pricks several times a day. The research involves collaboration with Google and German hardware company Infineon, which jointly developed a small radar device and sought input from select teams around the world on potential applications. Also read: ET's comprehensive diabetes page "We want to sense blood inside the body without actually having to sample any fluid. Our hope is this can be realised as a smartwatch to monitor glucose continuously," said George Shaker, an engineering professor at the University of Waterloo in Canada. The system at Waterloo uses the radar device to send high-frequency radio waves into liquids containing various levels of glucose and receive radio waves that are reflected back to it.

Nick Lane is an evolutionary biochemist at University College London who thinks about the big questions of life: how it began, how it is maintained, why we age and die, and why we have sex. Shunning the habit of our times to regard these as questions for evolutionary genetics, Lane insists that our fundamental biochemical mechanisms--particularly those through which living cells generate energy--may determine or limit these facts of life. Lane has been steadily constructing an alternative, complementary view of evolution to the one in which genes compete for reproductive success and survival. He has argued that some of the big shifts during evolutionary history, such as the appearance of complex cells called eukaryotes (like our own) and the emergence of multicellular life forms, are best understood by considering the energetic constraints. Lane's book Life Ascending: The Ten Great Inventions of Evolution was awarded the 2010 Royal Society Science Books Prize, the top prize in the United Kingdom for books on science. His 2015 book The Vital Question: Why Is Life the Way It Is? has been described as "game-changing" and "brimming with bold and important ideas." It offers a new, detailed model for how life might have begun by harnessing the incipient chemical energy at deep-sea vents. Bill Gates called The Vital Question "an amazing inquiry into the origins of life."