To enhance the efficiency of image-kernel convolution operations in convolutional neural networks, we introduce a Vector Systolic Array Accelerator with adaptable lane-width. This architecture utilizes multiple vector lanes for concurrent data element processing to address computational bottlenecks. To further improve the throughput, we propose a novel hybrid tiled vector systolic design in which we partition the hardware resources to efficiently utilize them along with a unique data mapping strategy. In this approach, we choose some tiles to use LUTs and others to use DSPs. We observe that the throughput of the vector systolic accelerator is 7x and 1.26x higher for single-tile and multi-tile configurations than their non-vector counterparts respectively. The hybrid tile design significantly increases tile count, achieving a competitive peak throughput of 1165 GOPs and 1072 GOPs for optimal lane width of Vector-6 and 8, which is 3.8x better than related work.

Attention, as a core layer of the ubiquitous Transformer architecture, is the bottleneck for large language models and long-context applications. elaborated an approach to speed up attention on GPUs through minimizing memory reads/writes. However, it has yet to take advantage of new capabilities present in recent hardware, with FlashAttention-2 achieving only 35% utilization on the H100 GPU. We develop three main techniques to speed up attention on Hopper GPUs: exploiting asynchrony of the Tensor Cores and TMA to (1) overlap overall computation and data movement via warp-specialization and (2) interleave block-wise matmul and softmax operations, and (3) block quantization and incoherent processing that leverages hardware support for FP8 low-precision. We demonstrate that our method, FlashAttention-3, achieves speedup on H100 GPUs by 1.5-2.0$\times$ with BF16 reaching up to 840 TFLOPs/s (85\% utilization), and with FP8 reaching 1.3 PFLOPs/s. We validate that FP8 FlashAttention-3 achieves 2.6$\times$ lower numerical error than a baseline FP8 attention.

Age is one of the major known risk factors for Alzheimer's Disease (AD). Detecting AD early is crucial for effective treatment and preventing irreversible brain damage. Brain age, a measure derived from brain imaging reflecting structural changes due to aging, may have the potential to identify AD onset, assess disease risk, and plan targeted interventions. Deep learning-based regression techniques to predict brain age from magnetic resonance imaging (MRI) scans have shown great accuracy recently. However, these methods are subject to an inherent regression to the mean effect, which causes a systematic bias resulting in an overestimation of brain age in young subjects and underestimation in old subjects. This weakens the reliability of predicted brain age as a valid biomarker for downstream clinical applications. Here, we reformulate the brain age prediction task from regression to classification to address the issue of systematic bias. Recognizing the importance of preserving ordinal information from ages to understand aging trajectory and monitor aging longitudinally, we propose a novel ORdinal Distance Encoded Regularization (ORDER) loss that incorporates the order of age labels, enhancing the model's ability to capture age-related patterns. Extensive experiments and ablation studies demonstrate that this framework reduces systematic bias, outperforms state-of-art methods by statistically significant margins, and can better capture subtle differences between clinical groups in an independent AD dataset. Our implementation is publicly available at https://github.com/jaygshah/Robust-Brain-Age-Prediction.

Harnessing the power of deep neural networks in the medical imaging domain is challenging due to the difficulties in acquiring large annotated datasets, especially for rare diseases, which involve high costs, time, and effort for annotation. Unsupervised disease detection methods, such as anomaly detection, can significantly reduce human effort in these scenarios. While anomaly detection typically focuses on learning from images of healthy subjects only, real-world situations often present unannotated datasets with a mixture of healthy and diseased subjects. Recent studies have demonstrated that utilizing such unannotated images can improve unsupervised disease and anomaly detection. However, these methods do not utilize knowledge specific to registered neuroimages, resulting in a subpar performance in neurologic disease detection. To address this limitation, we propose Brainomaly, a GAN-based image-to-image translation method specifically designed for neurologic disease detection. Brainomaly not only offers tailored image-to-image translation suitable for neuroimages but also leverages unannotated mixed images to achieve superior neurologic disease detection. Additionally, we address the issue of model selection for inference without annotated samples by proposing a pseudo-AUC metric, further enhancing Brainomaly's detection performance. Extensive experiments and ablation studies demonstrate that Brainomaly outperforms existing state-of-the-art unsupervised disease and anomaly detection methods by significant margins in Alzheimer's disease detection using a publicly available dataset and headache detection using an institutional dataset. The code is available from https://github.com/mahfuzmohammad/Brainomaly.

Stock market investors debate and heavily discuss stock ideas, investing strategies, news and market movements on social media platforms. The discussions are significantly longer in length and require extensive domain expertise for understanding. In this paper, we curate such discussions and construct a first-of-its-kind of abstractive summarization dataset. Our curated dataset consists of 7888 Reddit posts and manually constructed summaries for 400 posts. We robustly evaluate the summaries and conduct experiments on SOTA summarization tools to showcase their limitations. We plan to make the dataset publicly available. The sample dataset is available here: https://dhyeyjani.github.io/RSMC

Autograding systems are increasingly being deployed to meet the challenge of teaching programming at scale. We propose a methodology for extending autograders to provide meaningful feedback for incorrect programs. Our methodology starts with the instructor identifying the concepts and skills important to each programming assignment, designing the assignment, and designing a comprehensive test suite. Tests are then applied to code submissions to learn classes of common errors and produce classifiers to automatically categorize errors in future submissions. The instructor maps the errors to concepts and skills and writes hints to help students find their misconceptions and mistakes. We have applied the methodology to two assignments from our Introduction to Computer Science course. We used submissions from one semester of the class to build classifiers and write hints for observed common errors. We manually validated the automatic error categorization and potential usefulness of the hints using submissions from a second semester. We found that the hints given for erroneous submissions should be helpful for 96% or more of the cases. Based on these promising results, we have deployed our hints and are currently collecting submissions and feedback from students and instructors.

Mobile phones and tablets can be considered as the first incarnation of the post-PC era. Their explosive adoption rate has been driven by a number of factors, with the most signifcant influence being applications (apps) and app markets. Individuals and organizations are able to develop and publish apps, and the most popular form of monetization is mobile advertising.