Practical applications employing deep learning must guarantee inference quality. However, we found that the inference quality of state-of-the-art and state-of-the-practice in practical applications has a long tail distribution. In the real world, many tasks have strict requirements for the quality of deep learning inference, such as safety-critical and mission-critical tasks. The fluctuation of inference quality seriously affects its practical applications, and the quality at the tail may lead to severe consequences. State-of-the-art and state-of-the-practice with outstanding inference quality designed and trained under loose constraints still have poor inference quality under constraints with practical application significance. On the one hand, the neural network models must be deployed on complex systems with limited resources. On the other hand, safety-critical and mission-critical tasks need to meet more metric constraints while ensuring high inference quality. We coin a new term, ``tail quality,'' to characterize this essential requirement and challenge. We also propose a new metric, ``X-Critical-Quality,'' to measure the inference quality under certain constraints. This article reveals factors contributing to the failure of using state-of-the-art and state-of-the-practice algorithms and systems in real scenarios. Therefore, we call for establishing innovative methodologies and tools to tackle this enormous challenge.

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.

Most existing distillation methods ignore the flexible role of the temperature in the loss function and fix it as a hyper-parameter that can be decided by an inefficient grid search. In general, the temperature controls the discrepancy between two distributions and can faithfully determine the difficulty level of the distillation task. Keeping a constant temperature, i.e., a fixed level of task difficulty, is usually sub-optimal for a growing student during its progressive learning stages. In this paper, we propose a simple curriculum-based technique, termed Curriculum Temperature for Knowledge Distillation (CTKD), which controls the task difficulty level during the student's learning career through a dynamic and learnable temperature. Specifically, following an easy-to-hard curriculum, we gradually increase the distillation loss w.r.t. the temperature, leading to increased distillation difficulty in an adversarial manner. As an easy-to-use plug-in technique, CTKD can be seamlessly integrated into existing knowledge distillation frameworks and brings general improvements at a negligible additional computation cost. Extensive experiments on CIFAR-100, ImageNet-2012, and MS-COCO demonstrate the effectiveness of our method. Our code is available at https://github.com/zhengli97/CTKD.

Deep Metric Learning (DML) is a group of techniques that aim to measure the similarity between objects through the neural network. Although the number of DML methods has rapidly increased in recent years, most previous studies cannot effectively handle noisy data, which commonly exists in practical applications and often leads to serious performance deterioration. To overcome this limitation, in this paper, we build a connection between noisy samples and hard samples in the framework of self-paced learning, and propose a \underline{B}alanced \underline{S}elf-\underline{P}aced \underline{M}etric \underline{L}earning (BSPML) algorithm with a denoising multi-similarity formulation, where noisy samples are treated as extremely hard samples and adaptively excluded from the model training by sample weighting. Especially, due to the pairwise relationship and a new balance regularization term, the sub-problem \emph{w.r.t.} sample weights is a nonconvex quadratic function. To efficiently solve this nonconvex quadratic problem, we propose a doubly stochastic projection coordinate gradient algorithm. Importantly, we theoretically prove the convergence not only for the doubly stochastic projection coordinate gradient algorithm, but also for our BSPML algorithm. Experimental results on several standard data sets demonstrate that our BSPML algorithm has better generalization ability and robustness than the state-of-the-art robust DML approaches.

Image super-resolution is a common task on mobile and IoT devices, where one often needs to upscale and enhance low-resolution images and video frames. While numerous solutions have been proposed for this problem in the past, they are usually not compatible with low-power mobile NPUs having many computational and memory constraints. In this Mobile AI challenge, we address this problem and propose the participants to design an efficient quantized image super-resolution solution that can demonstrate a real-time performance on mobile NPUs. The participants were provided with the DIV2K dataset and trained INT8 models to do a high-quality 3X image upscaling. The runtime of all models was evaluated on the Synaptics VS680 Smart Home board with a dedicated edge NPU capable of accelerating quantized neural networks. All proposed solutions are fully compatible with the above NPU, demonstrating an up to 60 FPS rate when reconstructing Full HD resolution images. A detailed description of all models developed in the challenge is provided in this paper.

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.

几乎没有弹出的文本分类旨在在几个弹奏方案下对文本进行分类。以前的大多数方法都采用基于优化的元学习来获得任务分布。但是，由于少数样本和复杂模型之间的匹配以及有用的任务功能之间的区别，这些方法遭受了过度拟合问题的影响。为了解决这个问题，我们通过梯度相似性（AMGS）方法提出了一种新颖的自适应元学习器，以提高模型的泛化能力。具体而言，拟议的AMG基于两个方面缓解了过度拟合：（i）通过内部循环中的自我监督的辅助任务来获取样品的潜在语义表示并改善模型的概括，（ii）利用适应性元学习者通过适应性元学习者通过梯度通过相似性，可以在外环中基底学习者获得的梯度上增加约束。此外，我们对正则化对整个框架的影响进行系统分析。对几个基准测试的实验结果表明，与最先进的优化元学习方法相比，提出的AMG始终提高了很少的文本分类性能。

由于缺乏异常样品，因此仅具有正常样本的先验知识的异常检测才吸引更多的注意力。现有的基于CNN的像素重建方法遇到了两个问题。首先，重建源和目标是包含无法区分的语义信息的原始像素值。其次，CNN倾向于很好地重建正常样品和异常情况，使它们仍然很难区分。在本文中，我们提出异常检测变压器（ADTR）将变压器应用于重建预训练的特征。预训练的功能包含可区分的语义信息。同样，采用变压器限制以很好地重构异常，因此一旦重建失败，就可以轻松检测到异常。此外，我们提出了新的损失函数，使我们的方法与正常样本的情况以及具有图像级和像素级标记为异常的异常情况兼容。通过添加简单的合成或外部无关异常，可以进一步提高性能。广泛的实验是在包括MVTEC-AD和CIFAR-10在内的异常检测数据集上进行的。与所有基线相比，我们的方法取得了卓越的性能。

在图像之间生成健壮和可靠的对应关系是多种应用程序的基本任务。为了在全球和局部粒度上捕获上下文，我们提出了Aspanformer，这是一种基于变压器的无探测器匹配器，建立在层次的注意力结构上，采用了一种新颖的注意操作，能够以自适应方式调整注意力跨度。为了实现这一目标，首先，在每个跨注意阶段都会回归流图，以定位搜索区域的中心。接下来，在中心周围生成一个采样网格，其大小不是根据固定的经验配置为固定的，而是根据与流图一起估计的像素不确定性的自适应计算。最后，在派生区域内的两个图像上计算注意力，称为注意跨度。通过这些方式，我们不仅能够维持长期依赖性，而且能够在高相关性的像素之间获得细粒度的注意，从而补偿基本位置和匹配任务中的零件平滑度。在广泛的评估基准上的最新准确性验证了我们方法的强匹配能力。

基于深度学习的单图像超分辨率（SISR）方法引起了人们的关注，并在现代高级GPU上取得了巨大的成功。但是，大多数最先进的方法都需要大量参数，记忆和计算资源，这些参数通常会显示在当前移动设备CPU/NPU上时显示出较低的推理时间。在本文中，我们提出了一个简单的普通卷积网络，该网络具有快速最近的卷积模块（NCNET），该模块对NPU友好，可以实时执行可靠的超级分辨率。提出的最近的卷积具有与最近的UP采样相同的性能，但更快，更适合Android NNAPI。我们的模型可以很容易地在具有8位量化的移动设备上部署，并且与所有主要的移动AI加速器完全兼容。此外，我们对移动设备上的不同张量操作进行了全面的实验，以说明网络体系结构的效率。我们的NCNET在DIV2K 3X数据集上进行了训练和验证，并且与其他有效的SR方法的比较表明，NCNET可以实现高保真SR结果，同时使用更少的推理时间。我们的代码和预估计的模型可在\ url {https://github.com/algolzw/ncnet}上公开获得。