Optimizing for In-memory Deep Learning with Emerging Memory Technology
This addresses energy and performance bottlenecks for deep learning systems using emerging memory, but is incremental as it builds on existing in-memory computing paradigms.
The paper tackles the problem of accuracy loss in in-memory deep learning due to instability in emerging memory technology, proposing optimization techniques that recover state-of-the-art accuracy and achieve at least an order of magnitude higher energy efficiency.
In-memory deep learning computes neural network models where they are stored, thus avoiding long distance communication between memory and computation units, resulting in considerable savings in energy and time. In-memory deep learning has already demonstrated orders of magnitude higher performance density and energy efficiency. The use of emerging memory technology promises to increase the gains in density, energy, and performance even further. However, emerging memory technology is intrinsically unstable, resulting in random fluctuations of data reads. This can translate to non-negligible accuracy loss, potentially nullifying the gains. In this paper, we propose three optimization techniques that can mathematically overcome the instability problem of emerging memory technology. They can improve the accuracy of the in-memory deep learning model while maximizing its energy efficiency. Experiments show that our solution can fully recover most models' state-of-the-art accuracy, and achieves at least an order of magnitude higher energy efficiency than the state-of-the-art.