PhD Thesis

Happy to say that my PhD thesis—“Memristive Crossbars as Hardware Accelerators: Modelling, Design and New Uses”—has been made available online by UCL library services. You can find it (including full-text PDF) here.

As previously discussed in some of the blog posts1, my research focused on using memristive crossbar arrays to accelerate matrix operations, which are heavily utilized in fields like machine learning. I investigated (1) the negative effects of using nonideal analog devices, (2) ways to improve the performance of these devices and systems that they comprise, and (3) new ways of using memristive crossbar arrays.

  1. “The Power Crisis of Modern Computing”, “Analogue Solution to the Power-Hungry Field of Machine Learning”, “Memristive neural networks perform better when they work in teams”. ↩︎