物理内存计算：实时重建大脑皮层

An artificial neural network built into a computer memory chip reconstructs the human cortex with high accuracy in real time Xin Zheng1 and Ilia Valov1,2 C omputers predict physical processes by translat- ing them into mathematical equations, approxi- mating how numerical values change over an interval, and weighting the importance of specific features. Real-time computational simulations can closely replicate a phenomenon (1, 2). However, the to- tal time to complete computational tasks is longer than the time it takes for physical processes to occur (3). This latency is problematic if the goal is to computationally reconstruct the human brain’s cortical surface—the wrinkled, outermost layer of neural tissue—where electric pulses transmit infor- mation in tens of milliseconds intervals (4). …