A primer on stochastic computing and its challenges

Add to your list(s) Download to your calendar using vCal

• Julie Hsiao (University of Toronto)
• Monday 20 November 2023, 10:00-11:00
• Department of Electrical and Electronic Engineering, Room 909B.

If you have a question about this talk, please contact George A Constantinides.

In the present computing landscape, the significance of low-power and low-area circuit design has become increasingly evident to support emerging applications on IoT edge devices. We explore a non-traditional computing paradigm known as stochastic computing (SC). SC encodes and processes data as probabilities. Values are encoded in a bitstream format, where the ratio of 0s to 1s represents a probability value to be operated on. This encoding format allows computation to be carried out with extremely small units (e.g., multiplication is reduced to a single AND gate), enabling ultra low power circuits. However, unlike traditional binary computing, SC needs to handle data values with care to avoid the risk of unacceptably inaccurate computation results. In this talk, I’ll first give a primer on SC and then present some of our recent work on addressing two of the key challenges in SC: correlation and early termination.

This talk is part of the CAS Talks series.

This talk is included in these lists:

• Andrea Picciau's list
• CAS Talks
• Circuits and Systems Group: Internal Seminars
• Department of Electrical and Electronic Engineering, Room 909B

Note that ex-directory lists are not shown.