变形金刚已成为主要的机器学习工作负载，它们不仅是自然语言处理任务的事实上的标准，而且还将部署在其他领域，例如视觉和语音识别。许多基于变压器的应用程序都是实时系统，例如机器翻译和Web搜索。这些实时系统通常具有严格的端到端推理潜伏期需求。不幸的是，尽管大多数变压器计算都来自基质乘法，但变压器还包括几种非线性组件，它们在推理过程中倾向于成为瓶颈。在这项工作中，我们加快了张量流处理器上BERT模型的推断。通过小心地将所有非线性组件与矩阵乘法组件融合在一起，我们能够有效地利用芯片矩阵乘法单元，从而通过BERT-1通过BERT-1通过BERT-BASE，确定性的尾巴延迟为130 $ \ MU $ s，比当前的最新时间快6倍。

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

Accurate diagnosis of Alzheimer's disease (AD) is both challenging and time consuming. With a systematic approach for early detection and diagnosis of AD, steps can be taken towards the treatment and prevention of the disease. This study explores the practical application of deep learning models for diagnosis of AD. Due to computational complexity, large training times and limited availability of labelled dataset, a 3D full brain CNN (convolutional neural network) is not commonly used, and researchers often prefer 2D CNN variants. In this study, full brain 3D version of well-known 2D CNNs were designed, trained and tested for diagnosis of various stages of AD. Deep learning approach shows good performance in differentiating various stages of AD for more than 1500 full brain volumes. Along with classification, the deep learning model is capable of extracting features which are key in differentiating the various categories. The extracted features align with meaningful anatomical landmarks, that are currently considered important in identification of AD by experts. An ensemble of all the algorithm was also tested and the performance of the ensemble algorithm was superior to any individual algorithm, further improving diagnosis ability. The 3D versions of the trained CNNs and their ensemble have the potential to be incorporated in software packages that can be used by physicians/radiologists to assist them in better diagnosis of AD.

We present a framework for ranking images within their class based on the strength of spurious cues present. By measuring the gap in accuracy on the highest and lowest ranked images (we call this spurious gap), we assess spurious feature reliance for $89$ diverse ImageNet models, finding that even the best models underperform in images with weak spurious presence. However, the effect of spurious cues varies far more dramatically across classes, emphasizing the crucial, often overlooked, class-dependence of the spurious correlation problem. While most spurious features we observe are clarifying (i.e. improving test-time accuracy when present, as is typically expected), we surprisingly find many cases of confusing spurious features, where models perform better when they are absent. We then close the spurious gap by training new classification heads on lowly ranked (i.e. without common spurious cues) images, resulting in improved effective robustness to distribution shifts (ObjectNet, ImageNet-R, ImageNet-Sketch). We also propose a second metric to assess feature reliability, finding that spurious features are generally less reliable than non-spurious (core) ones, though again, spurious features can be more reliable for certain classes. To enable our analysis, we annotated $5,000$ feature-class dependencies over {\it all} of ImageNet as core or spurious using minimal human supervision. Finally, we show the feature discovery and spuriosity ranking framework can be extended to other datasets like CelebA and WaterBirds in a lightweight fashion with only linear layer training, leading to discovering a previously unknown racial bias in the Celeb-A hair classification.

Recommender systems are ubiquitous in most of our interactions in the current digital world. Whether shopping for clothes, scrolling YouTube for exciting videos, or searching for restaurants in a new city, the recommender systems at the back-end power these services. Most large-scale recommender systems are huge models trained on extensive datasets and are black-boxes to both their developers and end-users. Prior research has shown that providing recommendations along with their reason enhances trust, scrutability, and persuasiveness of the recommender systems. Recent literature in explainability has been inundated with works proposing several algorithms to this end. Most of these works provide item-style explanations, i.e., `We recommend item A because you bought item B.' We propose a novel approach, RecXplainer, to generate more fine-grained explanations based on the user's preference over the attributes of the recommended items. We perform experiments using real-world datasets and demonstrate the efficacy of RecXplainer in capturing users' preferences and using them to explain recommendations. We also propose ten new evaluation metrics and compare RecXplainer to six baseline methods.

Tasks critical to enterprise profitability, such as customer churn prediction, fraudulent account detection or customer lifetime value estimation, are often tackled by models trained on features engineered from customer data in tabular format. Application-specific feature engineering adds development, operationalization and maintenance costs over time. Recent advances in representation learning present an opportunity to simplify and generalize feature engineering across applications. When applying these advancements to tabular data researchers deal with data heterogeneity, variations in customer engagement history or the sheer volume of enterprise datasets. In this paper, we propose a novel approach to encode tabular data containing customer transactions, purchase history and other interactions into a generic representation of a customer's association with the business. We then evaluate these embeddings as features to train multiple models spanning a variety of applications. CASPR, Customer Activity Sequence-based Prediction and Representation, applies Transformer architecture to encode activity sequences to improve model performance and avoid bespoke feature engineering across applications. Our experiments at scale validate CASPR for both small and large enterprise applications.

视觉问题回答（VQA）是一项多模式的任务，涉及从输入图像中回答问题，以语义了解图像的内容并以自然语言回答。由于VQA系统回答的问题范围，使用VQA进行灾难管理是一项重要的研究。但是，主要的挑战是评估受影响地区的标签产生的延迟。为了解决这个问题，我们部署了预先训练的剪辑模型，该模型在视觉图像对中进行了训练。但是，我们从经验上看到该模型的零击性能差。因此，我们相反，我们使用此模型中的文本和图像的预训练嵌入，进行我们的监督培训，并超过Floodnet数据集上的先前最新结果。我们将其扩展到持续的设置，这是一种更现实的情况。我们解决了使用各种经验重播方法的灾难性遗忘的问题。我们的培训运行可在以下网址提供：https：//wandb.ai/compyle/continual_vqa_final

深度学习研究引起了广泛的兴趣，导致出现了各种各样的技术创新和应用。由于深度学习研究的很大比例关注基于视觉的应用，因此存在使用其中一些技术来实现低功率便携式医疗保健诊断支持解决方案的潜力。在本文中，我们提出了一个基于硬件的嵌入式软件实施显微镜诊断支持系统，用于POC案例研究：（a）厚血液涂片中的疟疾，（b）痰液样品中的结核病，以及（c）（c）粪便中的肠道寄生虫感染样品。我们使用基于挤压网络的模型来减少网络大小和计算时间。我们还利用训练有素的量化技术来进一步减少学习模型的记忆足迹。这使基于显微镜的病原体检测将实验室专家级别的精度分类为独立的嵌入式硬件平台。与基于CPU的常规实施相比，提议的实施功率更高6倍，并且推理时间为$ \ sim $ 3 ms/示例。

混合整数程序（MIP）通常通过分支结合算法解决。最近，学会模仿专家强的分支启发式的快速近似，由于它成功地减少了解决MIP的运行时间，因此引起了人们的关注。但是，现有的学习与分支方法假设整个培训数据都可以在一次培训中获得。这个假设通常不正确，如果随着时间的推移以连续的方式提供培训数据，现有技术会遭受灾难性遗忘。在这项工作中，我们研究了迄今未开发的终身学习范式，以在混合整数程序上分支。为了减轻灾难性的遗忘，我们提出了Limip，该limip是由以两部分图的形式对MIP实例进行建模的想法，我们使用双方图形注意力网络将其映射到嵌入式空间。这种丰富的嵌入空间避免了通过应用知识蒸馏和弹性重量巩固的灾难性遗忘，其中我们学习参数的关键是保持疗效，因此受到保护，免受明显的漂移。我们评估了一系列NP硬性问题的利润，并确定与现有基线相比，在面对终身学习时，Limip的速度高达50％。

深度学习网络已在各种应用中表现出高性能，例如图像分类，语音识别和自然语言处理。但是，存在使用对抗攻击所利用的主要漏洞。对抗性攻击通过稍微稍微更改输入图像，使其对肉眼几乎无法检测到图像，但导致网络的分类非常不同。本文探讨了使用两种类型的体系结构：MobileNetV3和Resnet50探讨图像分割DeepLabV3模型上预计的梯度下降（PGD）攻击和自适应面膜分割攻击（ASMA），发现PGD在更改分割方面非常一致它的目标虽然ASMA对多类目标的概括不那么有效。然而，这种攻击的存在使所有图像分类深度学习网络处于剥削的危险之中。