The vastness of the archival chemistry literature is both a blessing and a curse. The reaction that you're looking for is probably in there, provided you take enough time to search for it. Gao et al. trained a neural network model on 10 million known reactions to speed up this process. Specifically, the model was charged with predicting a catalyst, reagents, solvents, and temperature to achieve a given transformation. When tested, the model's top-10 list of suggestions produced a close match to actual conditions nearly 70% of the time, with a 20 C error margin in temperature.
This article was originally published on TechRepublic. Aerial imagery: Photos taken from the air, often with UAVs in smart farming. Used to assist farmers to determine the condition of a field. It is the integrated internal and external networking of farming operations as a result of the emergence of smart technology in agriculture. Agro-chemicals: Chemicals used in agriculture, which include fertilizers, herbicides, and pesticides.
Replanting trees after a wildfire or logging operation is an extremely labor intensive and expensive task. Carrying a bag of seedlings and using a dibble bar or shovel across steep debris-covered terrain can wear out a human. A new company, DroneSeed, has a solution. They are designing a system around a swarm of drones that can plant tree seeds in places where they have a decent chance of survival. First they survey the area with a drone using lidar and a multispectral camera to map the terrain and the vegetation.
In the mid-20th century, food production from agriculture sharply increased worldwide; however, this was achieved through heavy use of agrochemicals. Extensive collateral damage from excessive use of pesticides, herbicides, and fertilizers has occurred to the wider environment. This has led to biodiversity loss, pesticide resistance and the emergence of new pests, pollution and decline of freshwater supplies, and soil degradation and erosion, as well as direct harm to health. In a Review, Pretty examines the alternative approaches that can achieve sustainable intensification of farming systems by integrating pest management with agroecological systems to minimize costs, maximize yields, restore ecosystem services, and ensure environmental enhancement. The mid-20th century brought agricultural transformation and the "Green Revolution." New crop varieties and livestock breeds--combined with increased use of inorganic fertilizers, manufactured pesticides, and machinery--led to sharp increases in food production from agriculture worldwide. Yet this period of agricultural intensification was accompanied by considerable harm to the environment. This imposed costs on economies and made agricultural systems less efficient by degrading ecosystem goods and services. The desire for agriculture to produce more food without environmental harm, and even to make positive contributions to natural and social capital, has been reflected in many calls for more sustainable agriculture. Sustainable intensification (SI) comprises agricultural processes or systems in which production is maintained or increased while progressing toward substantial enhancement of environmental outcomes.
Oxygen on-board future space missions will be made from the recycled breath of astronauts. The Advanced Closed Loop System (ACLS) has been built by the European Space Agency (ESA) and is now being installed on-board the orbiting spacecraft. The apparatus recycles half the carbon dioxide (CO2) exhaled by astronauts and converts it into oxygen. Scientists have heralded the invention as an important step towards long-term missions to mars and beyond. ESA astronaut Alexander Gerst poses with the ACLS life-support rack, newly installed on the International Space Station.
In this work, we extend the SchNet architecture by using weighted skip connections to assemble the final representation. This enables us to study the relative importance of each interaction block for property prediction. We demonstrate on both the QM9 and MD17 dataset that their relative weighting depends strongly on the chemical composition and configurational degrees of freedom of the molecules which opens the path towards a more detailed understanding of machine learning models for molecules.
Science hasn't been giving us a tremendous amount of good news these days. We've screwed up the environment so badly, it's hard to even call it an environment anymore. And that's coming back to bite (or sting) us: Bee populations, which we rely on to pollinate our crops, are plummeting. But science is also coming to the rescue, by gluing QR codes to bumblebees' backs and tracking their movements with a robotic camera. Researchers have created a system that tracks individual bees as well as the dynamics of whole colonies exposed to imidacloprid, a neurotoxin that belongs to the infamous neonicotinoid group of pesticides.
Recently, locality sensitive hashing (LSH) was shown to be effective for MIPS and several algorithms including $L_2$-ALSH, Sign-ALSH and Simple-LSH have been proposed. In this paper, we introduce the norm-range partition technique, which partitions the original dataset into sub-datasets containing items with similar 2-norms and builds hash index independently for each sub-dataset. We prove that norm-range partition reduces the query processing complexity for all existing LSH based MIPS algorithms under mild conditions. The key to performance improvement is that norm-range partition allows to use smaller normalization factor most sub-datasets. For efficient query processing, we also formulate a unified framework to rank the buckets from the hash indexes of different sub-datasets. Experiments on real datasets show that norm-range partition significantly reduces the number of probed for LSH based MIPS algorithms when achieving the same recall.
Perfumers look out: IBM Research partnered up with one of the top producers of flavors and fragrances, Symrise, to create an perfume-concocting AI. Named Philyra, after the Greed goddess of fragrance, it uses machine learning to sift through thousands of ingredients, formulas and industry trends to derive what IBM considers to be unique combinations. IBM is leveraging the AI to help perfumers design the next great scent rather than a machine that will replace experts of the human nose. Philyra looks at thousands of formulas and raw materials to identify patterns and new combinations to find a potential gap in the market and fill it with a new scent. It finds alternative raw materials, deduces the dosage based on human usage patterns and how humans tend to respond before comparing it to existing fragrances.
Skilled perfumers bring art and science together to design new fragrances, a talent that takes ten or more years to develop. Crafting a fragrance that leaves an impression is one of the most important components a consumer considers when forming a positive or negative opinion about everyday products like laundry detergent, deodorant, air freshener and, of course, cologne and perfume. What if artificial intelligence (AI) could learn from these professionals to augment the process of developing new fragrances or identify completely novel creative pathways? With this in mind, my team at IBM Research, together with Symrise, one of the top global producers of flavors and fragrances, created an AI system that can learn about formulas, raw materials, historical success data and industry trends. Building on previous IBM research using AI to pair flavors and for recipe creation, as well as our new IBM Research AI for Product Composition, we created Philyra.