b'THE STATE OF RETAIL DIAMONDSJEWELRYCOLORED DESIGN STONESMACHINE LEARNING, EXPLAINEDTo talk about machine grading, it is useful to first understand the concept of machine learning and how it is positioned under the larger umbrella of artificial intelligence, or AI. AI refers to the field and development of computer systems that can do tasks that require human intelligence. Though they may not realize it, people utilize artificial intelligence every day, like when they are texting and their smartphone guesses which word theyre typing (known as predictive texting), or when their favorite streaming service recommends TV shows or movies they might like.Scientists developed the first AI programs in the 1950s, led by indi-viduals like British mathematician Alan Turing, creator of the famed Turing Test, and IBM computer scientist Arthur Samuel, who is widely credited with coining the phrase machine learning. Machine learning is a subset of AI that involves developing algorithms and statistical models that enable computer systems to learn from data without constantly having to be reprogrammed, explains Iu Ayala, found-er and CEO of data science consultancy firm Gradient Insight. These algorithms can identify patterns and make decisions based on the data thats fed into them. They can be a valuable tool for a wide range of tasks, including image recognition, language translation, fraud detection and diamond grading. THE FINAL FRONTIERFor three of the 4Cs, the GIA-developed measures of diamond quality that have been in use since the 1940s, using machines to grade Machines arediamonds is nothing new. more accurateCarat weight involves the simplest of ma-than people,chinesa scalewhile GIA invented the first system to measure proportions for cut grade, period. Theythe Proportion Scope, in 1967. The technology are definitelyhas evolved since then, including the introduc-more consistenttion of Sarines DiaMension system in 1992, [too].which automated the process.No one grades cut manually. Its all doneDAVID BLOCK,through technology, Sarine CEO DavidSARINEBlock says. This historical photo above shows Kenneth Moore, head of GIAs Gem Instruments division in the 1960s, using the first version of the Proportion Scope, an early machine used to help determine a diamonds cut TECHNOLOGIES Gemologists at GIA have been using thegrade, while the below photo is of a modern-day Gemological Institute of America grading lab.Colorimeter to grade color since the 1950s.(Both photosGIA)The technology has advanced to the point where the Colorimeter, which is used only in GIA labs, is able to gradeThe amount of data required for a machine to truly learn varies the vast majority of diamonds without human intervention. depending on the complexity of the task at hand, he says. In gener-In recent years, Sarine and GIA began investigating bringing so- al, the more data that is fed into a machine, the better it will be able called smart technology into the fold, particularly for clarity grading,to learn and make predictions. the element of diamond quality GIA Executive Vice President andFor Sarine, gathering the requisite amount of data meant opening Chief Laboratory and Research Officer Tom Moses has described asgrading labs, one in Israel and one in India, in 2018. the last frontier for machine grading.The company asked clients to send diamonds with GIA reports to the Block says Sarine developed the first automated color and claritylabs. It then cross-checked the machines grades against those of its own grading systems in 2016 while GIA started research with IBM onhuman graders as well as the GIA reports that accompanied each stone.automated clarity grading in earnest in 2017/2018.Block says over a two-year period, Sarine fed tens of thousands In order to utilize this type of automated technology, a companyof diamonds into its color and clarity machines, essentially teaching has to have a large quantity of data to feed into the machine, as ma- them what a D-color diamond looks like versus a G, H or I stone or, chines learn in much the same way people dothrough repetitionin the case of clarity, how a VS1 differs from a VS2, or an SI1 differs and memorization, Ayala, the AI expert, notes.from an SI2.Continued on page 5654 STATE OF THE MAJORS 2023'