我们将Kearns-Vazirani学习算法扩展到能够处理随时间变化的系统。我们提出了一种新的学习算法，该算法可以重复使用和更新以前学习的行为，在Learnlib库中实现它，并在大型示例中对其进行评估，我们在算法的两次运行中进行了少量调整。在这些实验中，我们的算法显着优于经典的Kearns-Vazirani学习算法和当前最新的自适应算法。

监视自动实时流处理系统的行为已成为现实世界应用中最相关的问题之一。这种系统的复杂性已在很大程度上依赖于高维输入数据和数据饥饿的机器学习（ML）算法。我们提出了一个灵活的系统，功能监视（FM），该系统在此类数据集中检测数据漂移，并具有较小且恒定的内存足迹和流应用程序中的小计算成本。该方法基于多变量统计测试，并且是由设计驱动的数据（从数据中估算了完整的参考分布）。它监视系统使用的所有功能，同时每当发生警报时提供可解释的功能排名（以帮助根本原因分析）。系统的计算和记忆轻度是由于使用指数移动直方图而导致的。在我们的实验研究中，我们用其参数分析了系统的行为，更重要的是显示了它检测到与单个特征无直接相关的问题的示例。这说明了FM如何消除添加自定义信号以检测特定类型问题的需求，并且监视功能可用空间通常足够。

Landing an unmanned aerial vehicle unmanned aerial vehicle (UAV) on top of an unmanned surface vehicle (USV) in harsh open waters is a challenging problem, owing to forces that can damage the UAV due to a severe roll and/or pitch angle of the USV during touchdown. To tackle this, we propose a novel model predictive control (MPC) approach enabling a UAV to land autonomously on a USV in these harsh conditions. The MPC employs a novel objective function and an online decomposition of the oscillatory motion of the vessel to predict, attempt, and accomplish the landing during near-zero tilt of the landing platform. The nonlinear prediction of the motion of the vessel is performed using visual data from an onboard camera. Therefore, the system does not require any communication with the USV or a control station. The proposed method was analyzed in numerous robotics simulations in harsh and extreme conditions and further validated in various real-world scenarios.

Language modeling, a central task in natural language processing, involves estimating a probability distribution over strings. In most cases, the estimated distribution sums to 1 over all finite strings. However, in some pathological cases, probability mass can ``leak'' onto the set of infinite sequences. In order to characterize the notion of leakage more precisely, this paper offers a measure-theoretic treatment of language modeling. We prove that many popular language model families are in fact tight, meaning that they will not leak in this sense. We also generalize characterizations of tightness proposed in previous works.

After just a few hundred training updates, a standard probabilistic model for language generation has likely not yet learnt many semantic or syntactic rules of natural language, which inherently makes it difficult to estimate the right probability distribution over next tokens. Yet around this point, these models have identified a simple, loss-minimising behaviour: to output the unigram distribution of the target training corpus. The use of such a crude heuristic raises the question: Rather than wasting precious compute resources and model capacity for learning this strategy at early training stages, can we initialise our models with this behaviour? Here, we show that we can effectively endow our model with a separate module that reflects unigram frequency statistics as prior knowledge. Standard neural language generation architectures offer a natural opportunity for implementing this idea: by initialising the bias term in a model's final linear layer with the log-unigram distribution. Experiments in neural machine translation demonstrate that this simple technique: (i) improves learning efficiency; (ii) achieves better overall performance; and (iii) appears to disentangle strong frequency effects, encouraging the model to specialise in non-frequency-related aspects of language.

In this work, we investigate the representation capacity of multilayer perceptron networks that use the sine as activation function - sinusoidal neural networks. We show that the layer composition in such networks compacts information. For this, we prove that the composition of sinusoidal layers expands as a sum of sines consisting of a large number of new frequencies given by linear combinations of the weights of the network's first layer. We provide the expression of the corresponding amplitudes in terms of the Bessel functions and give an upper bound for them that can be used to control the resulting approximation.

The usage of deep neural networks in safety-critical systems is limited by our ability to guarantee their correct behavior. Runtime monitors are components aiming to identify unsafe predictions and discard them before they can lead to catastrophic consequences. Several recent works on runtime monitoring have focused on out-of-distribution (OOD) detection, i.e., identifying inputs that are different from the training data. In this work, we argue that OOD detection is not a well-suited framework to design efficient runtime monitors and that it is more relevant to evaluate monitors based on their ability to discard incorrect predictions. We call this setting out-ofmodel-scope detection and discuss the conceptual differences with OOD. We also conduct extensive experiments on popular datasets from the literature to show that studying monitors in the OOD setting can be misleading: 1. very good OOD results can give a false impression of safety, 2. comparison under the OOD setting does not allow identifying the best monitor to detect errors. Finally, we also show that removing erroneous training data samples helps to train better monitors.

In this paper, we seek to measure how much information a component in a neural network could extract from the representations fed into it. Our work stands in contrast to prior probing work, most of which investigates how much information a model's representations contain. This shift in perspective leads us to propose a new principle for probing, the architectural bottleneck principle: In order to estimate how much information a given component could extract, a probe should look exactly like the component. Relying on this principle, we estimate how much syntactic information is available to transformers through our attentional probe, a probe that exactly resembles a transformer's self-attention head. Experimentally, we find that, in three models (BERT, ALBERT, and RoBERTa), a sentence's syntax tree is mostly extractable by our probe, suggesting these models have access to syntactic information while composing their contextual representations. Whether this information is actually used by these models, however, remains an open question.

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.

There is an increasing need in our society to achieve faster advances in Science to tackle urgent problems, such as climate changes, environmental hazards, sustainable energy systems, pandemics, among others. In certain domains like chemistry, scientific discovery carries the extra burden of assessing risks of the proposed novel solutions before moving to the experimental stage. Despite several recent advances in Machine Learning and AI to address some of these challenges, there is still a gap in technologies to support end-to-end discovery applications, integrating the myriad of available technologies into a coherent, orchestrated, yet flexible discovery process. Such applications need to handle complex knowledge management at scale, enabling knowledge consumption and production in a timely and efficient way for subject matter experts (SMEs). Furthermore, the discovery of novel functional materials strongly relies on the development of exploration strategies in the chemical space. For instance, generative models have gained attention within the scientific community due to their ability to generate enormous volumes of novel molecules across material domains. These models exhibit extreme creativity that often translates in low viability of the generated candidates. In this work, we propose a workbench framework that aims at enabling the human-AI co-creation to reduce the time until the first discovery and the opportunity costs involved. This framework relies on a knowledge base with domain and process knowledge, and user-interaction components to acquire knowledge and advise the SMEs. Currently,the framework supports four main activities: generative modeling, dataset triage, molecule adjudication, and risk assessment.