Sukru Burc Eryilmaz

Feiwen Zhu, Arkadiusz Nowaczynski, Rundong Li et al.

AlphaFold2 has been hailed as a breakthrough in protein folding. It can rapidly predict protein structures with lab-grade accuracy. However, its implementation does not include the necessary training code. OpenFold is the first trainable public reimplementation of AlphaFold. AlphaFold training procedure is prohibitively time-consuming, and gets diminishing benefits from scaling to more compute resources. In this work, we conducted a comprehensive analysis on the AlphaFold training procedure based on Openfold, identified that inefficient communications and overhead-dominated computations were the key factors that prevented the AlphaFold training from effective scaling. We introduced ScaleFold, a systematic training method that incorporated optimizations specifically for these factors. ScaleFold successfully scaled the AlphaFold training to 2080 NVIDIA H100 GPUs with high resource utilization. In the MLPerf HPC v3.0 benchmark, ScaleFold finished the OpenFold benchmark in 7.51 minutes, shown over $6\times$ speedup than the baseline. For training the AlphaFold model from scratch, ScaleFold completed the pretraining in 10 hours, a significant improvement over the seven days required by the original AlphaFold pretraining baseline.

Weier Wan, Rajkumar Kubendran, Clemens Schaefer et al.

Realizing today's cloud-level artificial intelligence functionalities directly on devices distributed at the edge of the internet calls for edge hardware capable of processing multiple modalities of sensory data (e.g. video, audio) at unprecedented energy-efficiency. AI hardware architectures today cannot meet the demand due to a fundamental "memory wall": data movement between separate compute and memory units consumes large energy and incurs long latency. Resistive random-access memory (RRAM) based compute-in-memory (CIM) architectures promise to bring orders of magnitude energy-efficiency improvement by performing computation directly within memory. However, conventional approaches to CIM hardware design limit its functional flexibility necessary for processing diverse AI workloads, and must overcome hardware imperfections that degrade inference accuracy. Such trade-offs between efficiency, versatility and accuracy cannot be addressed by isolated improvements on any single level of the design. By co-optimizing across all hierarchies of the design from algorithms and architecture to circuits and devices, we present NeuRRAM - the first multimodal edge AI chip using RRAM CIM to simultaneously deliver a high degree of versatility for diverse model architectures, record energy-efficiency $5\times$ - $8\times$ better than prior art across various computational bit-precisions, and inference accuracy comparable to software models with 4-bit weights on all measured standard AI benchmarks including accuracy of 99.0% on MNIST and 85.7% on CIFAR-10 image classification, 84.7% accuracy on Google speech command recognition, and a 70% reduction in image reconstruction error on a Bayesian image recovery task. This work paves a way towards building highly efficient and reconfigurable edge AI hardware platforms for the more demanding and heterogeneous AI applications of the future.

Sukru Burc Eryilmaz, Duygu Kuzum, Rakesh Jeyasingh et al.

Recent advances in neuroscience together with nanoscale electronic device technology have resulted in huge interests in realizing brain-like computing hardwares using emerging nanoscale memory devices as synaptic elements. Although there has been experimental work that demonstrated the operation of nanoscale synaptic element at the single device level, network level studies have been limited to simulations. In this work, we demonstrate, using experiments, array level associative learning using phase change synaptic devices connected in a grid like configuration similar to the organization of the biological brain. Implementing Hebbian learning with phase change memory cells, the synaptic grid was able to store presented patterns and recall missing patterns in an associative brain-like fashion. We found that the system is robust to device variations, and large variations in cell resistance states can be accommodated by increasing the number of training epochs. We illustrated the tradeoff between variation tolerance of the network and the overall energy consumption, and found that energy consumption is decreased significantly for lower variation tolerance.

Sercan Arik, Sukru Burc Eryilmaz, Adam Goldberg

As the number of publicly traded companies as well as the amount of their financial data grows rapidly, it is highly desired to have tracking, analysis, and eventually stock selections automated. There have been few works focusing on estimating the stock prices of individual companies. However, many of those have worked with very small number of financial parameters. In this work, we apply machine learning techniques to address automated stock picking, while using a larger number of financial parameters for individual companies than the previous studies. Our approaches are based on the supervision of prediction parameters using company fundamentals, time-series properties, and correlation information between different stocks. We examine a variety of supervised learning techniques and found that using stock fundamentals is a useful approach for the classification problem, when combined with the high dimensional data handling capabilities of support vector machine. The portfolio our system suggests by predicting the behavior of stocks results in a 3% larger growth on average than the overall market within a 3-month time period, as the out-of-sample test suggests.