Neural inference at the frontier of energy, space, and time.

Computing, since its inception, has been processor-centric, with memory separated from compute. Inspired by the organic brain and optimized for inorganic silicon, NorthPole is a neural inference architecture that blurs this boundary by eliminating off-chip memory, intertwining compute with memory on-chip, and appearing externally as an active memory chip. NorthPole is a low-precision, massively parallel, densely interconnected, energy-efficient, and spatial computing architecture with a co-optimized, high-utilization programming model. On the ResNet50 benchmark image classification network, relative to a graphics processing unit (GPU) that uses a comparable 12-nanometer technology process, NorthPole achieves a 25 times higher energy metric of frames per second (FPS) per watt, a 5 times higher space metric of FPS per transistor, and a 22 times lower time metric of latency. Similar results are reported for the Yolo-v4 detection network. NorthPole outperforms all prevalent architectures, even those that use more-advanced technology processes.

Convolutional networks for fast, energy-efficient neuromorphic computing.

Deep networks are now able to achieve human-level performance on a broad spectrum of recognition tasks. Independently, neuromorphic computing has now demonstrated unprecedented energy-efficiency through a new chip architecture based on spiking neurons, low precision synapses, and a scalable communication network. Here, we demonstrate that neuromorphic computing, despite its novel architectural primitives, can implement deep convolution networks that (i) approach state-of-the-art classification accuracy across eight standard datasets encompassing vision and speech, (ii) perform inference while preserving the hardware's underlying energy-efficiency and high throughput, running on the aforementioned datasets at between 1,200 and 2,600 frames/s and using between 25 and 275 mW (effectively >6,000 frames/s per Watt), and (iii) can be specified and trained using backpropagation with the same ease-of-use as contemporary deep learning. This approach allows the algorithmic power of deep learning to be merged with the efficiency of neuromorphic processors, bringing the promise of embedded, intelligent, brain-inspired computing one step closer.

Efficient and generalizable statistical models of shape and appearance for analysis of cardiac MRI.

We present a framework for the analysis of short axis cardiac MRI, using statistical models of shape and appearance. The framework integrates temporal and structural constraints and avoids common optimization problems inherent in such high dimensional models. The first contribution is the introduction of an algorithm for fitting 3D active appearance models (AAMs) on short axis cardiac MRI. We observe a 44-fold increase in fitting speed and a segmentation accuracy that is on par with Gauss-Newton optimization, one of the most widely used optimization algorithms for such problems. The second contribution involves an investigation on hierarchical 2D+time active shape models (ASMs), that integrate temporal constraints and simultaneously improve the 3D AAM based segmentation. We obtain encouraging results (endocardial/epicardial error 1.43+/-0.49 mm/1.51+/-0.48 mm) on 7980 short axis cardiac MR images acquired from 33 subjects. We have placed our dataset online, for the community to use and build upon.