Browser Not Supported9.3 A 40nm 4.81TFLOPS/W 8b Floating-Point Training Processor for Non-Sparse Neural Networks Using Shared Exponent Bias and 24-Way Fused Multiply-Add Tree | IEEE Conference Publication | IEEE Xplore

Notice: IEEE Xplore is unable to display this page because you may be using an incompatible or unsupported browser.

For the best experience, please upgrade to a newer, supported browser using the links below. If you cannot upgrade your browser, a temporary IEEE Xplore site is available for basic searches with links to full-text.