在课堂学习学习中，预计该模型将在保持以前课程的知识的同时，不断地学习新课程。这里的挑战在于保留该模型在功能空间中有效代表先前类的能力，同时调整其代表传入的新类。我们提出了两个基于蒸馏的目标，用于类增量学习，以利用特征空间的结构来维持以前的课程的准确性，并使学习新课程。在我们的第一个目标（称为跨空间聚类（CSC））中，我们建议使用先前模型的特征空间结构来表征优化的方向，这些方向可以最大程度地保留类 - 特定类的所有实例应集体优化，对，以及他们应该集体优化的人。除了最大程度地减少忘记之外，这种间接的鼓励模型将所有类的实例聚集在当前功能空间中，并引起牛群免疫的感觉，从而使班级的所有样本都可以将模型共同与遗忘班级共同打击模型。我们的第二个目标被称为受控转移（CT）从研究班间转移的研究的逐步学习。 CT明确近似于和条件，当前模型在逐步到达类和先验类之间的语义相似性上。这使模型可以学习类，以使其从相似的先前类中最大化正向转移，从而提高可塑性，并最大程度地减少不同先验类别的负向后转移，从而增强稳定性。我们在两个基准数据集上执行了广泛的实验，并在三种突出的课堂学习方法的顶部添加了我们的方法（CSCCT）。我们观察到各种实验环境的性能一致。

通过利用和适应到目前为止获得的知识，人类具有识别和区分他们不熟悉的实例的天生能力。重要的是，他们实现了这一目标，而不会在早期学习中恶化表现。受此启发，我们识别并制定了NCDWF的新的，务实的问题设置：新颖的类发现而无需忘记，哪个任务是机器学习模型从未标记的数据中逐步发现实例的新颖类别，同时在先前看到的类别上保持其性能。我们提出1）一种生成伪内表示的方法，该表示的代理（不再可用）标记的数据，从而减轻遗忘的遗忘，2）基于相互信息的正常化程序，可以增强对新型类别的无聊发现，而3）a 3）当测试数据包含所见类别和看不见的类别的实例时，简单的已知类标识符可以有助于广义推断。我们介绍了基于CIFAR-10，CIFAR-100和IMAGENET-1000的实验协议，以衡量知识保留和新型类发现之间的权衡。我们广泛的评估表明，现有的模型在确定新类别的同时灾难性地忘记了先前看到的类别，而我们的方法能够有效地在竞争目标之间平衡。我们希望我们的工作能够吸引对这个新确定的实用问题设定的进一步研究。

开放世界对象检测（OWOD）是一个具有挑战性的计算机视觉问题，其中任务是检测一组已知的对象类别，同时识别未知对象。此外，该模型必须逐步学习在下一个培训集中所知的新类。不同于标准对象检测，OWOD设置会对在潜在的未知物体上生成质量候选建议的质量挑战，将未知物体与背景中的未知物体分开并检测不同的未知物体。在这里，我们介绍了一种新的基于端到端的变换器的框架OW-DETR，用于开放世界对象检测。建议的OW-DETR包括三个专用组成部分，即注意力驱动的伪标签，新颖性分类和对象评分，以明确地解决上述OWOD挑战。我们的OW-DETR明确地编码了多尺度上下文信息，具有较少的归纳偏差，使得从已知类传输到未知类，并且可以更好地区分未知对象和背景之间。综合实验是对两个基准进行的：MS-Coco和Pascal VOC。广泛的消融揭示了我们拟议的贡献的优点。此外，我们的模型优于最近引入的OWOD方法矿石，绝对增益在MS-Coco基准测试中的未知召回方面的1.8％至3.3％。在增量对象检测的情况下，OW-DETR以Pascal VOC基准上的所有设置优于最先进的。我们的代码和模型将公开发布。

在真实世界的环境中，可以通过对象检测器连续遇到来自新类的对象实例。当现有的对象探测器应用于这种情况时，它们在旧课程上的性能显着恶化。据报道，一些努力解决了这个限制，所有这些限制适用于知识蒸馏的变体，以避免灾难性的遗忘。我们注意到虽然蒸馏有助于保留以前的学习，但它阻碍了对新任务的快速适应性，这是增量学习的关键要求。在这种追求中，我们提出了一种学习方法，可以学习重塑模型梯度，使得跨增量任务的信息是最佳的共享。这可通过META学习梯度预处理来确保无缝信息传输，可最大限度地减少遗忘并最大化知识传输。与现有的元学习方法相比，我们的方法是任务不可知，允许将新类的增量添加到对象检测的高容量模型中。我们在Pascal-VOC和MS Coco Datasets上定义的各种增量学习设置中评估了我们的方法，我们的方法对最先进的方法进行了好评。

As various city agencies and mobility operators navigate toward innovative mobility solutions, there is a need for strategic flexibility in well-timed investment decisions in the design and timing of mobility service regions, i.e. cast as "real options" (RO). This problem becomes increasingly challenging with multiple interacting RO in such investments. We propose a scalable machine learning based RO framework for multi-period sequential service region design & timing problem for mobility-on-demand services, framed as a Markov decision process with non-stationary stochastic variables. A value function approximation policy from literature uses multi-option least squares Monte Carlo simulation to get a policy value for a set of interdependent investment decisions as deferral options (CR policy). The goal is to determine the optimal selection and timing of a set of zones to include in a service region. However, prior work required explicit enumeration of all possible sequences of investments. To address the combinatorial complexity of such enumeration, we propose a new variant "deep" RO policy using an efficient recurrent neural network (RNN) based ML method (CR-RNN policy) to sample sequences to forego the need for enumeration, making network design & timing policy tractable for large scale implementation. Experiments on multiple service region scenarios in New York City (NYC) shows the proposed policy substantially reduces the overall computational cost (time reduction for RO evaluation of > 90% of total investment sequences is achieved), with zero to near-zero gap compared to the benchmark. A case study of sequential service region design for expansion of MoD services in Brooklyn, NYC show that using the CR-RNN policy to determine optimal RO investment strategy yields a similar performance (0.5% within CR policy value) with significantly reduced computation time (about 5.4 times faster).

A "heart attack" or myocardial infarction (MI), occurs when an artery supplying blood to the heart is abruptly occluded. The "gold standard" method for imaging MI is Cardiovascular Magnetic Resonance Imaging (MRI), with intravenously administered gadolinium-based contrast (late gadolinium enhancement). However, no "gold standard" fully automated method for the quantification of MI exists. In this work, we propose an end-to-end fully automatic system (MyI-Net) for the detection and quantification of MI in MRI images. This has the potential to reduce the uncertainty due to the technical variability across labs and inherent problems of the data and labels. Our system consists of four processing stages designed to maintain the flow of information across scales. First, features from raw MRI images are generated using feature extractors built on ResNet and MoblieNet architectures. This is followed by the Atrous Spatial Pyramid Pooling (ASPP) to produce spatial information at different scales to preserve more image context. High-level features from ASPP and initial low-level features are concatenated at the third stage and then passed to the fourth stage where spatial information is recovered via up-sampling to produce final image segmentation output into: i) background, ii) heart muscle, iii) blood and iv) scar areas. New models were compared with state-of-art models and manual quantification. Our models showed favorable performance in global segmentation and scar tissue detection relative to state-of-the-art work, including a four-fold better performance in matching scar pixels to contours produced by clinicians.

We propose an approach for semantic imitation, which uses demonstrations from a source domain, e.g. human videos, to accelerate reinforcement learning (RL) in a different target domain, e.g. a robotic manipulator in a simulated kitchen. Instead of imitating low-level actions like joint velocities, our approach imitates the sequence of demonstrated semantic skills like "opening the microwave" or "turning on the stove". This allows us to transfer demonstrations across environments (e.g. real-world to simulated kitchen) and agent embodiments (e.g. bimanual human demonstration to robotic arm). We evaluate on three challenging cross-domain learning problems and match the performance of demonstration-accelerated RL approaches that require in-domain demonstrations. In a simulated kitchen environment, our approach learns long-horizon robot manipulation tasks, using less than 3 minutes of human video demonstrations from a real-world kitchen. This enables scaling robot learning via the reuse of demonstrations, e.g. collected as human videos, for learning in any number of target domains.

By transferring knowledge from large, diverse, task-agnostic datasets, modern machine learning models can solve specific downstream tasks either zero-shot or with small task-specific datasets to a high level of performance. While this capability has been demonstrated in other fields such as computer vision, natural language processing or speech recognition, it remains to be shown in robotics, where the generalization capabilities of the models are particularly critical due to the difficulty of collecting real-world robotic data. We argue that one of the keys to the success of such general robotic models lies with open-ended task-agnostic training, combined with high-capacity architectures that can absorb all of the diverse, robotic data. In this paper, we present a model class, dubbed Robotics Transformer, that exhibits promising scalable model properties. We verify our conclusions in a study of different model classes and their ability to generalize as a function of the data size, model size, and data diversity based on a large-scale data collection on real robots performing real-world tasks. The project's website and videos can be found at robotics-transformer.github.io

Large-scale data is an essential component of machine learning as demonstrated in recent advances in natural language processing and computer vision research. However, collecting large-scale robotic data is much more expensive and slower as each operator can control only a single robot at a time. To make this costly data collection process efficient and scalable, we propose Policy Assisted TeleOperation (PATO), a system which automates part of the demonstration collection process using a learned assistive policy. PATO autonomously executes repetitive behaviors in data collection and asks for human input only when it is uncertain about which subtask or behavior to execute. We conduct teleoperation user studies both with a real robot and a simulated robot fleet and demonstrate that our assisted teleoperation system reduces human operators' mental load while improving data collection efficiency. Further, it enables a single operator to control multiple robots in parallel, which is a first step towards scalable robotic data collection. For code and video results, see https://clvrai.com/pato

In this work, we demonstrate the offline FPGA realization of both recurrent and feedforward neural network (NN)-based equalizers for nonlinearity compensation in coherent optical transmission systems. First, we present a realization pipeline showing the conversion of the models from Python libraries to the FPGA chip synthesis and implementation. Then, we review the main alternatives for the hardware implementation of nonlinear activation functions. The main results are divided into three parts: a performance comparison, an analysis of how activation functions are implemented, and a report on the complexity of the hardware. The performance in Q-factor is presented for the cases of bidirectional long-short-term memory coupled with convolutional NN (biLSTM + CNN) equalizer, CNN equalizer, and standard 1-StpS digital back-propagation (DBP) for the simulation and experiment propagation of a single channel dual-polarization (SC-DP) 16QAM at 34 GBd along 17x70km of LEAF. The biLSTM+CNN equalizer provides a similar result to DBP and a 1.7 dB Q-factor gain compared with the chromatic dispersion compensation baseline in the experimental dataset. After that, we assess the Q-factor and the impact of hardware utilization when approximating the activation functions of NN using Taylor series, piecewise linear, and look-up table (LUT) approximations. We also show how to mitigate the approximation errors with extra training and provide some insights into possible gradient problems in the LUT approximation. Finally, to evaluate the complexity of hardware implementation to achieve 400G throughput, fixed-point NN-based equalizers with approximated activation functions are developed and implemented in an FPGA.