RESUMO
Implementations of artificial neural networks that borrow analogue techniques could potentially offer low-power alternatives to fully digital approaches1-3. One notable example is in-memory computing based on crossbar arrays of non-volatile memories4-7 that execute, in an analogue manner, multiply-accumulate operations prevalent in artificial neural networks. Various non-volatile memories-including resistive memory8-13, phase-change memory14,15 and flash memory16-19-have been used for such approaches. However, it remains challenging to develop a crossbar array of spin-transfer-torque magnetoresistive random-access memory (MRAM)20-22, despite the technology's practical advantages such as endurance and large-scale commercialization5. The difficulty stems from the low resistance of MRAM, which would result in large power consumption in a conventional crossbar array that uses current summation for analogue multiply-accumulate operations. Here we report a 64 × 64 crossbar array based on MRAM cells that overcomes the low-resistance issue with an architecture that uses resistance summation for analogue multiply-accumulate operations. The array is integrated with readout electronics in 28-nanometre complementary metal-oxide-semiconductor technology. Using this array, a two-layer perceptron is implemented to classify 10,000 Modified National Institute of Standards and Technology digits with an accuracy of 93.23 per cent (software baseline: 95.24 per cent). In an emulation of a deeper, eight-layer Visual Geometry Group-8 neural network with measured errors, the classification accuracy improves to 98.86 per cent (software baseline: 99.28 per cent). We also use the array to implement a single layer in a ten-layer neural network to realize face detection with an accuracy of 93.4 per cent.
RESUMO
We examined the effects of a 6-week 40-m one-way sprint interval training program (based on sprint time). 13 untrained healthy male collegiate students performed six 40-m sprints with a 60-s resting interval between sprints during the first week, and one sprint was added each week until the sixth week. If the 40-m sprint time exceeded 110% of the fastest baseline 40-m sprint time, the run was repeated. Repeated-sprint cycling test (every 3 weeks), quadriceps moment (every 2 weeks), and abdominal and thigh subcutaneous tissue thickness (every 2 weeks) were measured. Compared to baseline, mean power output improved at week 3 (16.27 vs. 17.73 Watt/kg, p=0.004). Regardless of side, quadriceps moment began to increase at week 4 (2.88 vs. 3.15 N·m/kg, p=0.03). Subcutaneous tissue thickness was reduced at week 2 (abdominal: 11.19 vs. 9.65 mm, p=0.01; thigh: 9.17 vs. 8.12 mm, p=0.009). Our results suggest that (1) sprint training with an intensity of 110% of the fastest baseline 40-m sprint time with the addition of one sprint per week produces similar effects to other training programs, and (2) untrained individuals need 4 weeks of training for strength development in the quadriceps and 2 weeks for reduction in fat tissue thickness.
