如果通常激励有能力的AI代理来寻求为我们指定的目标服务的权力，那么除了巨大的利益外，这些系统还将带来巨大的风险。在完全可观察到的环境中，大多数奖励功能都具有最佳的政策，该政策通过保持期权开放并保持活力来寻求权力。但是，现实世界既不是完全可观察到的，也不是代理人绝对最佳的。我们考虑了一系列的AI决策模型，从最佳，随机到通过学习和与环境互动所告知的选择。我们发现许多决策功能都是可以重新定位的，并且可重新定位的性足以引起寻求权力的趋势。我们的功能标准简单而广泛。我们表明，一系列定性决策程序激励代理寻求权力。我们通过在蒙特祖玛的报仇中推理了学到的政策激励措施来证明结果的灵活性。这些结果表明安全风险：最终，高度可重新定位的培训程序可能会训练寻求对人类权力的现实世界代理商。

AI目标通常很难正确指定。有些方法通过规范AI的副作用来解决此问题：代理必须用不完美的代理目标来权衡“他们造成了多少混乱”。我们通过援助游戏框架提出了一个正式的副作用正规化标准。在这些游戏中，代理解决了一个可观察到的马尔可夫决策过程（POMDP），代表了其对其应优化目标函数的不确定性。我们考虑在以后的时间步骤向代理揭示真正目标的设置。我们证明，通过将代理人奖励与代理商实现一系列未来任务的能力进行交易，可以解决此POMDP。我们通过在两个环境环境中的地面真相评估来证明问题形式化的合理性。

事实证明，关系决策树的合奏模型（行李和梯度提升）被证明是概率逻辑模型（PLM）领域中最有效的学习方法之一。尽管有效，但他们失去了PLM的最重要方面之一 - 可解释性。在本文中，我们考虑将大量博学的树木压缩成单个可解释的模型的问题。为此，我们提出了COTE（树的压缩），该Cote将单个小型决策列表作为压缩表示形式。Cote首先将树木转换为决策清单，然后借助原始训练集执行组合和压缩。实验评估证明了COTE在几个基准关系数据集中的有效性。

注意图是解释图像分类卷积网络的决策的流行方式。通常，对于感兴趣的每个图像，产生单一的注意图，其基于它们对分类的重要性分配给像素的权重。然而，单一的注意图提供了不完整的理解，因为通常有许多其他地图可以同样解释分类。在本文中，我们介绍了结构化的注意图（SAG），它通过捕获图像区域的不同组合影响分类器的信心来紧凑地代表图像的注意力映射。我们提出了一种方法来计算SAG和SAG的可视化，以便可以获得更深层次的洞察力进入分类器的决定。我们进行用户学习比较使用SAG对传统注意图的使用，以应对图像分类的反事实问题。我们的结果表明，当基于SAG与基线相比，用户在回答基于落下的比较反事实问题时更为正确。

一些研究人员推测智能强化学习（RL）代理商将被激励寻求资源和追求目标的权力。其他研究人员指出，RL代理商不需要具有人类的寻求技能本能。为了澄清这一讨论，我们开展了最优政策统计趋势的第一个正式理论。在马尔可夫决策过程的背景下，我们证明某些环境对称是足以实现对环境寻求权力的最佳政策。这些对称存在于许多环境中，其中代理可以关闭或销毁。我们证明，在这些环境中，大多数奖励功能使其通过保持一系列可用的选项来寻求电力，并在最大限度地提高平均奖励时，通过导航到更大的潜在终端状态。

This paper proposes a novel self-supervised based Cut-and-Paste GAN to perform foreground object segmentation and generate realistic composite images without manual annotations. We accomplish this goal by a simple yet effective self-supervised approach coupled with the U-Net based discriminator. The proposed method extends the ability of the standard discriminators to learn not only the global data representations via classification (real/fake) but also learn semantic and structural information through pseudo labels created using the self-supervised task. The proposed method empowers the generator to create meaningful masks by forcing it to learn informative per-pixel as well as global image feedback from the discriminator. Our experiments demonstrate that our proposed method significantly outperforms the state-of-the-art methods on the standard benchmark datasets.

Indian e-commerce industry has evolved over the last decade and is expected to grow over the next few years. The focus has now shifted to turnaround time (TAT) due to the emergence of many third-party logistics providers and higher customer expectations. The key consideration for delivery providers is to balance their overall operating costs while meeting the promised TAT to their customers. E-commerce delivery partners operate through a network of facilities whose strategic locations help to run the operations efficiently. In this work, we identify the locations of hubs throughout the country and their corresponding mapping with the distribution centers. The objective is to minimize the total network costs with TAT adherence. We use Genetic Algorithm and leverage business constraints to reduce the solution search space and hence the solution time. The results indicate an improvement of 9.73% in TAT compliance compared with the current scenario.

Independence testing is a fundamental and classical statistical problem that has been extensively studied in the batch setting when one fixes the sample size before collecting data. However, practitioners often prefer procedures that adapt to the complexity of a problem at hand instead of setting sample size in advance. Ideally, such procedures should (a) allow stopping earlier on easy tasks (and later on harder tasks), hence making better use of available resources, and (b) continuously monitor the data and efficiently incorporate statistical evidence after collecting new data, while controlling the false alarm rate. It is well known that classical batch tests are not tailored for streaming data settings, since valid inference after data peeking requires correcting for multiple testing, but such corrections generally result in low power. In this paper, we design sequential kernelized independence tests (SKITs) that overcome such shortcomings based on the principle of testing by betting. We exemplify our broad framework using bets inspired by kernelized dependence measures such as the Hilbert-Schmidt independence criterion (HSIC) and the constrained-covariance criterion (COCO). Importantly, we also generalize the framework to non-i.i.d. time-varying settings, for which there exist no batch tests. We demonstrate the power of our approaches on both simulated and real data.

This paper describes a simple yet efficient repetition-based modular system for speeding up air-traffic controllers (ATCos) training. E.g., a human pilot is still required in EUROCONTROL's ESCAPE lite simulator (see https://www.eurocontrol.int/simulator/escape) during ATCo training. However, this need can be substituted by an automatic system that could act as a pilot. In this paper, we aim to develop and integrate a pseudo-pilot agent into the ATCo training pipeline by merging diverse artificial intelligence (AI) powered modules. The system understands the voice communications issued by the ATCo, and, in turn, it generates a spoken prompt that follows the pilot's phraseology to the initial communication. Our system mainly relies on open-source AI tools and air traffic control (ATC) databases, thus, proving its simplicity and ease of replicability. The overall pipeline is composed of the following: (1) a submodule that receives and pre-processes the input stream of raw audio, (2) an automatic speech recognition (ASR) system that transforms audio into a sequence of words; (3) a high-level ATC-related entity parser, which extracts relevant information from the communication, i.e., callsigns and commands, and finally, (4) a speech synthesizer submodule that generates responses based on the high-level ATC entities previously extracted. Overall, we show that this system could pave the way toward developing a real proof-of-concept pseudo-pilot system. Hence, speeding up the training of ATCos while drastically reducing its overall cost.

In recent years the importance of Smart Healthcare cannot be overstated. The current work proposed to expand the state-of-art of smart healthcare in integrating solutions for Obsessive Compulsive Disorder (OCD). Identification of OCD from oxidative stress biomarkers (OSBs) using machine learning is an important development in the study of OCD. However, this process involves the collection of OCD class labels from hospitals, collection of corresponding OSBs from biochemical laboratories, integrated and labeled dataset creation, use of suitable machine learning algorithm for designing OCD prediction model, and making these prediction models available for different biochemical laboratories for OCD prediction for unlabeled OSBs. Further, from time to time, with significant growth in the volume of the dataset with labeled samples, redesigning the prediction model is required for further use. The whole process requires distributed data collection, data integration, coordination between the hospital and biochemical laboratory, dynamic machine learning OCD prediction mode design using a suitable machine learning algorithm, and making the machine learning model available for the biochemical laboratories. Keeping all these things in mind, Accu-Help a fully automated, smart, and accurate OCD detection conceptual model is proposed to help the biochemical laboratories for efficient detection of OCD from OSBs. OSBs are classified into three classes: Healthy Individual (HI), OCD Affected Individual (OAI), and Genetically Affected Individual (GAI). The main component of this proposed framework is the machine learning OCD prediction model design. In this Accu-Help, a neural network-based approach is presented with an OCD prediction accuracy of 86 percent.