Follow us twitter facebook
Edition: Brazil
Click here to subscribe toour free weekly newsletter click here
Science, R&D

Google Research team used machine learning to train AI to recognize smells

Last week, a Google Research team published a paper about how they trained an artificial intelligence (AI) using machine learning to reasonably predict "the relationship between a molecule’s structure and its odor," aka recognize different smells based on molecule shapes.

Predicting the relationship between a molecule’s structure and its odor remains a difficult. This problem, termed quantitative structure-odor relationship (QSOR) modeling, is an important challenge in chemistry, with a strong potential impact on the manufacture of synthetic fragrance, among others.

A team of Google researchers recently published a paper on how they trained (...)

A team of Google researchers recently published a paper on how they trained a neural network to connect a molecule’s structure with its odor. (Photo: © grinvalds/Istock.com)

A paper published on October 23, 2019, by a team of Google researchers tackles how machine learning can be used to find the relationship - which scientists have been trying to quantify for over 70 years - between a molecule’s odor and its structure.

While molecular structures can give scientists insight as to what something looks like or what it sounds like, "predicting the relationship between a molecule’s structure and its odor remains a difficult, decades-old task."

These scientists used machine learning to find this underlying quantitative relationship in a way reminiscent of how deep learning is used to predict visual and auditory characteristics.

Using a dataset of 5030 molecules labeled with odor descriptors (like fruity, bready, nutty and cheese) by olfactory experts (like professional perfumers), the team trained a Graph Neural Network to predict these labels based on the molecule’s shape. The perceptual and structural similarities of both local and global components of a molecule aided this neural network in recognizing which parts are responsible for which fragrance nodes.

According to the paper, being able to use machine learning to predict odors based on their molecular structures would "aid in the discovery of new synthetic odorants" so that natural products don’t need to be harvested in such high numbers.

Brazil Beauty News with AFP/Relaxnews

© 2019 - Brazil Beauty News - www.brazilbeautynews.com

latest news
Focus
MakeUp in New York moves to Javits Center to continue growing

MakeUp in New York moves to Javits Center to continue growing

The latest edition of MakeUp in NewYork was a huge success, announced the organisers. Held at the Center415 on 5th Avenue, in Manhattan, on September 11&12, 2019, the two-day B2B event dedicated to the makeup and skincare supply chain welcomed more than 3800 visitors, with a growth of 12% of international guests. Brazil, (...)

read more
Experts’ views
Fragrance Trends: The impact of the mineral

Aliénor Massenet
Fragrance Trends: The impact of the mineral

A solid, inorganic and crystallized substance, minerals are derived from a natural geological process. In perfume- making, minerality is a subjective concept because there is no mineral family strictly speaking. With wet and salty facets, mineral notes give perfumers a new way of expressing freshness in their creations. Perfumer (...)

read more

Features