Approximate Computing is a modern design paradigm that aims to exploit the inherent tolerance to errors in a wide set of applications. Providing good enough results in applications such as image processing and financial analysis, where a unique or golden answer does not exist, this paradigm improve performance in terms of execution time, area, and power/energy, even by orders of magnitude. This can be achieved, for instance, by skipping non-critical computations at software level, or reducing circuit complexity and lowering the operation voltage at hardware level.
In this seminar we are interested in studying the applicability of Approximate Computing in all the abstraction layers, from operative system and applications to architecture and arithmetic components, keeping a particular interest in the cross-layer interaction of approximations and the development of methodologies and tools to implement Approximate Computing.
Application: Registration form.
Meeting Schedule: Will be informed vie email after registration.