Issue link: https://iconnect007.uberflip.com/i/1424540
50 SMT007 MAGAZINE I NOVEMBER 2021 As most manufacturers now scramble around for materials, machines, and spare parts, as well as skilled and experienced labor, remem- ber that the most modern soware from ma- chine vendors, MES providers, and others are likely to be far more advanced than you real- ize, which reflects on the ability to cope much more easily with the business-related chang- es and limitations that are being forced upon us all. e argument that any change, includ- ing that of new soware, brings its own vari- ations and challenges, as well as thinking that it may be too late now as the global challenges are already with us, are not valid. e amount of control and visibility that the latest so- ware provides is at least an order of magnitude greater than the work and effort to introduce the soware. As for global challenges coming to an end, many think that is unrealistic, both in terms of dealing with ongoing consequences of what has happened already—for example, with the pandemic—as well as worsening factors such as climate change and potentially political is- sues, which are forcing rapid change in gov- ernment policy, risk management, and cus- tomer demand choices. e worst and the best are yet to come. e most modern so- ware is your friend and ally as we face the fu- ture together. SMT007 Michael Ford is the senior director of emerging industry strategy for Aegis Software. To read past columns or contact Ford, click here. Researchers led by the Institute of Scientific and Industrial Research (SANKEN) at Osaka University have trained a deep neural network to correctly de- termine the output state of quantum bits, despite environmental noise. The team's novel approach may allow quantum computers to become much more widely used. Modern computers are based on binary logic, in which each bit is constrained to be either a 1 or a 0. But thanks to the weird rules of quantum me- chanics, new experimental systems can achieve in- creased computing power by allowing quantum bits, also called qubits, to be in "superpositions" of 1 and 0. For example, the spins of electrons confined to tiny islands called quantum dots can be oriented both up and down simultaneously. However, when the final state of a bit is read out, it reverts to the classical behavior of being one orientation or the other. To make quantum computing reliable enough for consumer use, new systems will need to be cre- ated that can accurately record the output of each qubit even if there is a lot of noise in the signal. Now, a team of scientists led by SANKEN used a machine learning method called a deep neural net- work to discern the signal created by the spin ori- entation of electrons on quantum dots. "We devel- oped a classifier based on deep neural network to precisely measure a qubit state even with noisy sig- nals," co-author Takafumi Fujita explains. In the experimental system, only electrons with a particular spin orientation can leave a quantum dot. When this happens, a temporary "blip" of in- creased voltage is created. The team trained the machine learning algorithm to pick out these sig- nals from among the noise. The deep neural net- work they used had a convolutional neural network to identify the important signal features, combined with a recurrent neural network to monitor the time- series data. (Source: Osaka University) Cutting Through the Noise: AI Enables High-fidelity Quantum Computing