Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision | |||
|
start [2019/02/27 22:47] Andreas Moshovos |
start [2019/02/27 22:55] (current) Andreas Moshovos [Deep Learning Acceleration] |
||
|---|---|---|---|
| Line 12: | Line 12: | ||
| - | **Value-Based Acceleration: ** We are developing, designing and demonstrating a novel class of hardware accelerators for Deep Learning networks whose key feature is that they are **value-based**. Conventional accelerators rely mostly on the structure of computation, that is, which calculations are performed and how they communicate. Value-based accelerators further boost performance by taking advantage of expected properties in the runtime calculated value stream, such as, dynamically redundant or ineffectual computations, or the distribution of values, or even their bit content. In short, our accelerator designs, reduce the amount of work that needs to be performed for existing neural models and do so transparently to the model designer. | + | **Value-Based Acceleration: ** We are developing methods that reduce the work, storage and communication needed when executing Deep Learning models. We target optimizations at the middleware software and at the hardware levels so that they benefit out-of-the-box models and do not require intervention from the Machine Learning expert: developing models is hard enough already. Our methods rely on value properties exhibited by typical models such as value- and bit-sparsity and data type need variability. Our methods, however, do reward model optimizations. For example, our methods reward quantization to smaller data widths where possible but will still provide benefits for non-quantized models. Similarly for sparsity. |
| - | **Why are we pursuing these designs?** Because Deep Learning is transforming our world by leaps and bounds. One of the three drivers behind Deep Learning success is the computing hardware that enabled its first practical applications. While algorithmic improvements will allow Deep Learning to evolve, much hinges on hardware’s ability to keep delivering ever higher performance and data processing storage and processing capability. As Dennard scaling has seized, the only viable way to do so is by architecture specialization. | + | See overview articles here: [[https://ieeexplore.ieee.org/document/8364645|Exploiting Typical Values to Accelerate Deep Learning]] |