语言的自动处理在我们的生活中普遍存在，经常在我们的决策中扮演核心角色，例如为我们的消息和邮件选择措辞，翻译我们的读物，甚至与我们进行完整的对话。单词嵌入是现代自然语言处理系统的关键组成部分。它们提供了一种词的表示，从而提高了许多应用程序的性能，从而是含义的表现。单词嵌入似乎可以捕捉到原始文本中单词的含义的外观，但与此同时，它们还提炼了刻板印象和社会偏见，后来传达给最终应用。这样的偏见可能是歧视性的。检测和减轻这些偏见，以防止自动化过程的歧视行为非常重要，因为它们的规模可能比人类更有害。目前，有许多工具和技术可以检测和减轻单词嵌入中的偏见，但是它们为没有技术技能的人的参与带来了许多障碍。碰巧的是，大多数偏见专家，无论是社会科学家还是对偏见有害，没有这样的技能的环境，并且由于技术障碍而无法参与偏见检测过程。我们研究了现有工具中的障碍，并与不同种类的用户探索了它们的可能性和局限性。通过此探索，我们建议开发一种专门旨在降低技术障碍的工具，并提供探索能力，以满足愿意审核这些技术的专家，科学家和一般人的要求。

Binary Neural Networks (BNNs) are showing tremendous success on realistic image classification tasks. Notably, their accuracy is similar to the state-of-the-art accuracy obtained by full-precision models tailored to edge devices. In this regard, BNNs are very amenable to edge devices since they employ 1-bit to store the inputs and weights, and thus, their storage requirements are low. Also, BNNs computations are mainly done using xnor and pop-counts operations which are implemented very efficiently using simple hardware structures. Nonetheless, supporting BNNs efficiently on mobile CPUs is far from trivial since their benefits are hindered by frequent memory accesses to load weights and inputs. In BNNs, a weight or an input is stored using one bit, and aiming to increase storage and computation efficiency, several of them are packed together as a sequence of bits. In this work, we observe that the number of unique sequences representing a set of weights is typically low. Also, we have seen that during the evaluation of a BNN layer, a small group of unique sequences is employed more frequently than others. Accordingly, we propose exploiting this observation by using Huffman Encoding to encode the bit sequences and then using an indirection table to decode them during the BNN evaluation. Also, we propose a clustering scheme to identify the most common sequences of bits and replace the less common ones with some similar common sequences. Hence, we decrease the storage requirements and memory accesses since common sequences are encoded with fewer bits. We extend a mobile CPU by adding a small hardware structure that can efficiently cache and decode the compressed sequence of bits. We evaluate our scheme using the ReAacNet model with the Imagenet dataset. Our experimental results show that our technique can reduce memory requirement by 1.32x and improve performance by 1.35x.

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

Survival modeling in healthcare relies on explainable statistical models; yet, their underlying assumptions are often simplistic and, thus, unrealistic. Machine learning models can estimate more complex relationships and lead to more accurate predictions, but are non-interpretable. This study shows it is possible to estimate hospitalization for congestive heart failure by a 30 seconds single-lead electrocardiogram signal. Using a machine learning approach not only results in greater predictive power but also provides clinically meaningful interpretations. We train an eXtreme Gradient Boosting accelerated failure time model and exploit SHapley Additive exPlanations values to explain the effect of each feature on predictions. Our model achieved a concordance index of 0.828 and an area under the curve of 0.853 at one year and 0.858 at two years on a held-out test set of 6,573 patients. These results show that a rapid test based on an electrocardiogram could be crucial in targeting and treating high-risk individuals.

Bayesian optimization (BO) is one of the most effective methods for closed-loop experimental design and black-box optimization. However, a key limitation of BO is that it is an inherently sequential algorithm (one experiment is proposed per round) and thus cannot directly exploit high-throughput (parallel) experiments. Diverse modifications to the BO framework have been proposed in the literature to enable exploitation of parallel experiments but such approaches are limited in the degree of parallelization that they can achieve and can lead to redundant experiments (thus wasting resources and potentially compromising performance). In this work, we present new parallel BO paradigms that exploit the structure of the system to partition the design space. Specifically, we propose an approach that partitions the design space by following the level sets of the performance function and an approach that exploits partially-separable structures of the performance function found. We conduct extensive numerical experiments using a reactor case study to benchmark the effectiveness of these approaches against a variety of state-of-the-art parallel algorithms reported in the literature. Our computational results show that our approaches significantly reduce the required search time and increase the probability of finding a global (rather than local) solution.

我们研究不同损失功能对医学图像病变细分的影响。尽管在处理自然图像时，跨凝结（CE）损失是最受欢迎的选择，但对于生物医学图像分割，由于其处理不平衡的情况，软骰子损失通常是首选的。另一方面，这两个功能的组合也已成功地应用于此类任务中。一个较少研究的问题是在存在分布（OOD）数据的情况下所有这些损失的概括能力。这是指在测试时间出现的样本，这些样本是从与训练图像不同的分布中得出的。在我们的情况下，我们将模型训练在始终包含病变的图像上，但是在测试时间我们也有无病变样品。我们通过全面的实验对内窥镜图像和糖尿病脚图像的溃疡分割进行了全面的实验，分析了不同损失函数对分布性能的最小化对分布性能的影响。我们的发现令人惊讶：在处理OOD数据时，CE-DICE损失组合在分割分配图像中表现出色，这使我们建议通过这种问题采用CE损失，因为它的稳健性和能够概括为OOD样品。可以在\ url {https://github.com/agaldran/lesion_losses_ood}找到与我们实验相关的代码。

糖尿病性视网膜病变（DR）是发达国家工人衰老人群中失明的主要原因之一，这是由于糖尿病的副作用降低了视网膜的血液供应。深度神经网络已被广泛用于自动化系统中，以在眼底图像上进行DR分类。但是，这些模型需要大量带注释的图像。在医疗领域，专家的注释昂贵，乏味且耗时。结果，提供了有限数量的注释图像。本文提出了一种半监督的方法，该方法利用未标记的图像和标记的图像来训练一种检测糖尿病性视网膜病的模型。提出的方法通过自我监督的学习使用无监督的预告片，然后使用一小部分标记的图像和知识蒸馏来监督微调，以提高分类任务的性能。在Eyepacs测试和Messidor-2数据集中评估了此方法，仅使用2％的Eyepacs列车标记图像，分别使用0.94和0.89 AUC。

核密度估计（KDE）是使用最广泛的非参数密度估计方法之一。它是一种基于内存的方法，即它将整个培训数据集用于预测，这使其不适合大多数当前的大数据应用程序。已经提出了几种策略，例如基于树或基于哈希的估计器，以提高内核密度估计方法的效率。新型密度内核密度估计方法（DMKDE）使用密度矩阵，量子机械形式主义和随机傅立叶特征（显式内核近似）来产生密度估计。该方法的根源在KDE中，可以被视为近似方法，而无需基于内存的限制。在本文中，我们系统地评估了新型DMKDE算法，并将其与其他最新的快速程序进行比较，以近似于不同合成数据集的内核密度估计方法。我们的实验结果表明，在高维数据上执行时，显示了DMKDE与其竞争对手的计算密度估计和优势相提并论。我们将所有代码作为开源软件存储库提供。

如今有大量在线文档数据源。缺乏结构和格式之间的差异是自动​​从中提取信息的主要困难，这也对其使用和重复使用产生负面影响。在生物医学领域中，出现了DISNET平台，以通过大规模的异质来源为研究人员提供资源，以获取人类疾病网络范围的信息。具体来说，在该领域中，不仅提供从不同来源提取的信息，而且提供支持它的证据至关重要。本文提出了Eboca，该本体论描述了（i）生物医学领域概念及其之间的关联，以及（ii）支持这些关联的证据；目的是提供一个模式来改善该领域中的证据和生物医学关联的出版和描述。已成功评估了本体论，以确保没有错误，建模陷阱，并符合先前定义的功能要求。来自文本的一部分和自动关联提取子集的测试数据已根据所提出的本体论进行了转换，以创建可用于实际场景中的知识图，并且还用于评估所述本体论。

密度估计是统计和机器学习应用中的基本任务。内核密度估计是低维度非参数密度估计的强大工具；但是，其性能在更高的维度上很差。此外，其预测复杂性量表与更多的培训数据点线性线性。本文提出了一种神经密度估计的方法，可以看作是一种核密度估计的一种，但没有高预测计算复杂性。该方法基于密度矩阵，一种用于量子力学的形式主义和自适应傅立叶特征。可以在没有优化的情况下对该方法进行培训，但也可以与深度学习体系结构集成并使用梯度下降进行训练。因此，它可以看作是神经密度估计方法的一种形式。该方法在不同的合成和实际数据集中进行了评估，其性能与最新的神经密度估计方法进行了比较，从而获得了竞争结果。