持续学习研究的主要重点领域是通过设计新算法对分布变化更强大的新算法来减轻神经网络中的“灾难性遗忘”问题。尽管持续学习文献的最新进展令人鼓舞，但我们对神经网络的特性有助于灾难性遗忘的理解仍然有限。为了解决这个问题，我们不关注持续的学习算法，而是在这项工作中专注于模型本身，并研究神经网络体系结构对灾难性遗忘的“宽度”的影响，并表明宽度在遗忘遗产方面具有出人意料的显着影响。为了解释这种效果，我们从各个角度研究网络的学习动力学，例如梯度正交性，稀疏性和懒惰的培训制度。我们提供了与不同架构和持续学习基准之间的经验结果一致的潜在解释。

In lifelong learning, the learner is presented with a sequence of tasks, incrementally building a data-driven prior which may be leveraged to speed up learning of a new task. In this work, we investigate the efficiency of current lifelong approaches, in terms of sample complexity, computational and memory cost. Towards this end, we first introduce a new and a more realistic evaluation protocol, whereby learners observe each example only once and hyper-parameter selection is done on a small and disjoint set of tasks, which is not used for the actual learning experience and evaluation. Second, we introduce a new metric measuring how quickly a learner acquires a new skill. Third, we propose an improved version of GEM (Lopez-Paz & Ranzato, 2017), dubbed Averaged GEM (A-GEM), which enjoys the same or even better performance as GEM, while being almost as computationally and memory efficient as EWC and other regularizationbased methods. Finally, we show that all algorithms including A-GEM can learn even more quickly if they are provided with task descriptors specifying the classification tasks under consideration. Our experiments on several standard lifelong learning benchmarks demonstrate that A-GEM has the best trade-off between accuracy and efficiency. 1

Incremental learning (IL) has received a lot of attention recently, however, the literature lacks a precise problem definition, proper evaluation settings, and metrics tailored specifically for the IL problem. One of the main objectives of this work is to fill these gaps so as to provide a common ground for better understanding of IL. The main challenge for an IL algorithm is to update the classifier whilst preserving existing knowledge. We observe that, in addition to forgetting, a known issue while preserving knowledge, IL also suffers from a problem we call intransigence, inability of a model to update its knowledge. We introduce two metrics to quantify forgetting and intransigence that allow us to understand, analyse, and gain better insights into the behaviour of IL algorithms. We present RWalk, a generalization of EWC++ (our efficient version of EWC [7]) and Path Integral [26] with a theoretically grounded KL-divergence based perspective. We provide a thorough analysis of various IL algorithms on MNIST and CIFAR-100 datasets. In these experiments, RWalk obtains superior results in terms of accuracy, and also provides a better trade-off between forgetting and intransigence.

and widely used information measurement metric, particularly popularized for SSVEP- based Brain-Computer (BCI) interfaces. By combining speed and accuracy into a single-valued parameter, this metric aids in the evaluation and comparison of various target identification algorithms across different BCI communities. To accurately depict performance and inspire an end-to-end design for futuristic BCI designs, a more thorough examination and definition of ITR is therefore required. We model the symbiotic communication medium, hosted by the retinogeniculate visual pathway, as a discrete memoryless channel and use the modified capacity expressions to redefine the ITR. We use graph theory to characterize the relationship between the asymmetry of the transition statistics and the ITR gain with the new definition, leading to potential bounds on data rate performance. On two well-known SSVEP datasets, we compared two cutting-edge target identification methods. Results indicate that the induced DM channel asymmetry has a greater impact on the actual perceived ITR than the change in input distribution. Moreover, it is demonstrated that the ITR gain under the new definition is inversely correlated with the asymmetry in the channel transition statistics. Individual input customizations are further shown to yield perceived ITR performance improvements. An algorithm is proposed to find the capacity of binary classification and further discussions are given to extend such results to ensemble techniques.We anticipate that the results of our study will contribute to the characterization of the highly dynamic BCI channel capacities, performance thresholds, and improved BCI stimulus designs for a tighter symbiosis between the human brain and computer systems while enhancing the efficiency of the underlying communication resources.

Safe and smooth robot motion around obstacles is an essential skill for autonomous robots, especially when operating around people and other robots. Conventionally, due to real-time operation requirements and onboard computation limitations, many robot motion planning and control methods follow a two-step approach: first construct a (e.g., piecewise linear) collision-free reference path for a simplified robot model, and then execute the reference plan via path-following control for a more accurate and complex robot model. A challenge of such a decoupled robot motion planning and control method for highly dynamic robotic systems is ensuring the safety of path-following control as well as the successful completion of the reference plan. In this paper, we introduce a novel dynamical systems approach for online closed-loop time parametrization, called $\textit{a time governor}$, of a reference path for provably correct and safe path-following control based on feedback motion prediction, where the safety of robot motion under path-following control is continuously monitored using predicted robot motion. After introducing the general framework of time governors for safe path following, we present an example application for the fully actuated high-order robot dynamics using proportional-and-higher-order-derivative (PhD) path-following control whose feedback motion prediction is performed by Lyapunov ellipsoids and Vandemonde simplexes. In numerical simulations, we investigate the role of reference position and velocity feedback, and motion prediction on path-following performance and robot motion.

Context-sensitive two-point layer 5 pyramidal cells (L5PCs) were discovered as long ago as 1999. However, the potential of this discovery to provide useful neural computation has yet to be demonstrated. Here we show for the first time how a transformative L5PCs-driven deep neural network (DNN), termed the multisensory cooperative computing (MCC) architecture, can effectively process large amounts of heterogeneous real-world audio-visual (AV) data, using far less energy compared to best available 'point' neuron-driven DNNs. A novel highly-distributed parallel implementation on a Xilinx UltraScale+ MPSoC device estimates energy savings up to 245759 $ \times $ 50000 $\mu$J (i.e., 62% less than the baseline model in a semi-supervised learning setup) where a single synapse consumes $8e^{-5}\mu$J. In a supervised learning setup, the energy-saving can potentially reach up to 1250x less (per feedforward transmission) than the baseline model. The significantly reduced neural activity in MCC leads to inherently fast learning and resilience against sudden neural damage. This remarkable performance in pilot experiments demonstrates the embodied neuromorphic intelligence of our proposed cooperative L5PC that receives input from diverse neighbouring neurons as context to amplify the transmission of most salient and relevant information for onward transmission, from overwhelmingly large multimodal information utilised at the early stages of on-chip training. Our proposed approach opens new cross-disciplinary avenues for future on-chip DNN training implementations and posits a radical shift in current neuromorphic computing paradigms.

作为一个简单且强大的移动机器人基础，可以将差速器驱动器机器人建模为运动学独轮车，在工业和国内环境中找到了物流和服务机器人技术的重要应用。安全的机器人在障碍物周围导航是这样的独轮车机器人在复杂的混乱环境中执行各种有用任务的重要技能，尤其是在人和其他机器人周围。在本文中，作为标准圆形Lyapunov级集的更准确的替代方法，我们介绍了新型的锥形反馈运动预测方法，用于在标准的Unicycle运动控制方法下界定运动学Unicycle机器人机器人模型的近环运动轨迹。我们介绍了使用参考调速器的安全机器人导航的Unicycle反馈运动预测的应用，在该机器人的安全下，根据预测的机器人运动，不断监视独轮车运动的安全性。我们研究了运动预测对机器人行为在数值模拟中的作用，并得出结论，准确的反馈运动预测是安全和快速机器人导航的关键。

人类活动识别是计算机视觉中的新出现和重要领域，旨在确定个体或个体正在执行的活动。该领域的应用包括从体育中生成重点视频到智能监视和手势识别。大多数活动识别系统依赖于卷积神经网络（CNN）的组合来从数据和复发性神经网络（RNN）中进行特征提取来确定数据的时间依赖性。本文提出并设计了两个用于人类活动识别的变压器神经网络：一个经常性变压器（RET），这是一个专门的神经网络，用于对数据序列进行预测，以及视觉变压器（VIT），一种用于提取显着的变压器的变压器（VIT）图像的特征，以提高活动识别的速度和可扩展性。我们在速度和准确性方面提供了对拟议的变压器神经网络与现代CNN和基于RNN的人类活动识别模型的广泛比较。

基于视觉的人类活动识别已成为视频分析领域的重要研究领域之一。在过去的十年中，已经引入了许多先进的深度学习算法，以识别视频流中复杂的人类行为。这些深度学习算法对人类活动识别任务显示出令人印象深刻的表现。但是，这些新引入的方法仅专注于模型性能或这些模型在计算效率和鲁棒性方面的有效性，从而导致其解决挑战性人类活动识别问题的提议中的偏差折衷。为了克服当代深度学习模型对人类活动识别的局限性，本文提出了一个计算高效但通用的空间级联框架，该框架利用了深层歧视性的空间和时间特征，以识别人类活动的识别。为了有效地表示人类行动，我们提出了有效的双重注意卷积神经网络（CNN）体系结构，该结构利用统一的通道空间注意机制来提取视频框架中以人为中心的显着特征。双通道空间注意力层与卷积层一起学会在具有特征图数量的物体的空间接收场中更加专注。然后将提取的判别显着特征转发到堆叠的双向封闭式复发单元（BI-GRU），以使用前进和后传球梯度学习，以实现长期时间建模和对人类行为的识别。进行了广泛的实验，其中获得的结果表明，与大多数当代动作识别方法相比，所提出的框架的执行时间的改善最高167倍。

准确且一致的边界分割在肿瘤体积估计及其在医学图像分割领域中的处理中起着重要作用。在全球范围内，肺癌是死亡的主要原因之一，肺结节的早期发现对于早期癌症诊断和患者的存活率至关重要。这项研究的目的是证明DeepHealth Toolkit的可行性，包括PYECVL和PYEDDL库（包括精确的肺结节）。使用PYECVL和PYEDDL在UnitoChest上进行了肺结节分割的实验，以进行数据预处理以及神经网络训练。结果描述了在较宽的直径范围内对肺结节的准确分割，并且在传统检测方法上的准确性更好。本文中使用的数据集和代码可作为基线参考公开提供。