在终生学习中，代理人在整个生命中都在不重复的一生中学习，就像人类一样，在不断变化的环境中。因此，终身学习带来了许多研究问题，例如连续领域的转移，这导致了非平稳的奖励和环境动态。由于其连续的性质，这些非平稳性很难检测和应对。因此，需要探索策略和学习方法，这些方法能够跟踪稳定的领域变化并适应它们。我们提出反应性探索，以跟踪和反应终生增强学习中持续的域转移，并相应地更新策略。为此，我们进行实验以研究不同的勘探策略。我们从经验上表明，政策阶级家族的代表更适合终身学习，因为它们比Q学习更快地适应了分销的变化。因此，政策梯度方法从反应性探索中获利最大，并在终身学习中显示出良好的结果，并进行了持续的领域变化。我们的代码可在以下网址提供：https：//github.com/ml-jku/reactive-ecploration。

在部分可观察到的马尔可夫决策过程（POMDP）中，代理通常使用过去的表示来近似基础MDP。我们建议利用冷冻验证的语言变压器（PLT）进行病史表示和压缩，以提高样品效率。为了避免对变压器进行训练，我们引入了Frozenhopfield，该菲尔德自动将观察结果与预处理的令牌嵌入相关联。为了形成这些关联，现代的Hopfield网络存储了这些令牌嵌入，这些嵌入是通过查询获得的查询来检索的，这些嵌入者通过随机但固定的观察结果获得。我们的新方法Helm，启用了Actor-Critic网络体系结构，该架构包含用于历史记录表示的历史模块的审计语言变压器。由于不需要学习过去的代表，因此掌舵比竞争对手要高得多。在Miligrid和Procgen环境上，Helm掌舵取得了新的最新结果。我们的代码可在https://github.com/ml-jku/helm上找到。

在现实世界中，通过弱势政策影响环境可能是昂贵的或非常危险的，因此妨碍了现实世界的加强学习应用。离线强化学习（RL）可以从给定数据集中学习策略，而不与环境进行交互。但是，数据集是脱机RL算法的唯一信息源，并确定学习策略的性能。我们仍然缺乏关于数据集特征如何影响不同离线RL算法的研究。因此，我们对数据集特性如何实现离散动作环境的离线RL算法的性能的全面实证分析。数据集的特点是两个度量：（1）通过轨迹质量（TQ）测量的平均数据集返回和（2）由状态 - 动作覆盖（SACO）测量的覆盖范围。我们发现，禁止政策深度Q网家族的变体需要具有高SACO的数据集来表现良好。将学习策略朝向给定数据集的算法对具有高TQ或SACO的数据集进行了良好。对于具有高TQ的数据集，行为克隆优先级或类似于最好的离线RL算法。

剪辑在零拍传输学习任务上产生了令人印象深刻的结果，并被视为BERT或GPT3等基础模型。具有丰富表示形式的剪辑视觉模型是使用Infonce目标和自然语言监督对特定任务进行微调之前进行预训练的。尽管剪辑在零拍传输学习方面表现出色，但它遭受了解释的问题，也就是说，它的重点是一个或几个功能，同时忽略了其他相关功能。该问题是由于原始多模式数据中未充分提取协方差结构而引起的。我们建议使用现代Hopfield网络来解决解释的问题。他们检索到的嵌入具有富集的协方差结构，该结构源自存储嵌入中特征的共发生。但是，现代的Hopfield网络增加了阻碍学习的Infonce目标的饱和效应。我们建议使用Infoloob目标来减轻这种饱和效果。我们介绍了小说``对比抛弃了一个增压'（Cloob），该小说使用现代的Hopfield网络与Infoloob Opportions一起进行协方差丰富。在实验中，我们将Cloob与概念标题进行预培训后的剪辑和YFCC数据集进行了比较，相对于其在其他数据集上的零拍传输学习性能。 Cloob在所有考虑的架构和数据集中始终在零摄像转移学习上胜过剪辑。

Due to the environmental impacts caused by the construction industry, repurposing existing buildings and making them more energy-efficient has become a high-priority issue. However, a legitimate concern of land developers is associated with the buildings' state of conservation. For that reason, infrared thermography has been used as a powerful tool to characterize these buildings' state of conservation by detecting pathologies, such as cracks and humidity. Thermal cameras detect the radiation emitted by any material and translate it into temperature-color-coded images. Abnormal temperature changes may indicate the presence of pathologies, however, reading thermal images might not be quite simple. This research project aims to combine infrared thermography and machine learning (ML) to help stakeholders determine the viability of reusing existing buildings by identifying their pathologies and defects more efficiently and accurately. In this particular phase of this research project, we've used an image classification machine learning model of Convolutional Neural Networks (DCNN) to differentiate three levels of cracks in one particular building. The model's accuracy was compared between the MSX and thermal images acquired from two distinct thermal cameras and fused images (formed through multisource information) to test the influence of the input data and network on the detection results.

Temporal action segmentation tags action labels for every frame in an input untrimmed video containing multiple actions in a sequence. For the task of temporal action segmentation, we propose an encoder-decoder-style architecture named C2F-TCN featuring a "coarse-to-fine" ensemble of decoder outputs. The C2F-TCN framework is enhanced with a novel model agnostic temporal feature augmentation strategy formed by the computationally inexpensive strategy of the stochastic max-pooling of segments. It produces more accurate and well-calibrated supervised results on three benchmark action segmentation datasets. We show that the architecture is flexible for both supervised and representation learning. In line with this, we present a novel unsupervised way to learn frame-wise representation from C2F-TCN. Our unsupervised learning approach hinges on the clustering capabilities of the input features and the formation of multi-resolution features from the decoder's implicit structure. Further, we provide the first semi-supervised temporal action segmentation results by merging representation learning with conventional supervised learning. Our semi-supervised learning scheme, called ``Iterative-Contrastive-Classify (ICC)'', progressively improves in performance with more labeled data. The ICC semi-supervised learning in C2F-TCN, with 40% labeled videos, performs similar to fully supervised counterparts.

We propose Panoptic Lifting, a novel approach for learning panoptic 3D volumetric representations from images of in-the-wild scenes. Once trained, our model can render color images together with 3D-consistent panoptic segmentation from novel viewpoints. Unlike existing approaches which use 3D input directly or indirectly, our method requires only machine-generated 2D panoptic segmentation masks inferred from a pre-trained network. Our core contribution is a panoptic lifting scheme based on a neural field representation that generates a unified and multi-view consistent, 3D panoptic representation of the scene. To account for inconsistencies of 2D instance identifiers across views, we solve a linear assignment with a cost based on the model's current predictions and the machine-generated segmentation masks, thus enabling us to lift 2D instances to 3D in a consistent way. We further propose and ablate contributions that make our method more robust to noisy, machine-generated labels, including test-time augmentations for confidence estimates, segment consistency loss, bounded segmentation fields, and gradient stopping. Experimental results validate our approach on the challenging Hypersim, Replica, and ScanNet datasets, improving by 8.4, 13.8, and 10.6% in scene-level PQ over state of the art.

Terabytes of data are collected every day by wind turbine manufacturers from their fleets. The data contain valuable real-time information for turbine health diagnostics and performance monitoring, for predicting rare failures and the remaining service life of critical parts. And yet, this wealth of data from wind turbine fleets remains inaccessible to operators, utility companies, and researchers as manufacturing companies prefer the privacy of their fleets' turbine data for business strategic reasons. The lack of data access impedes the exploitation of opportunities, such as improving data-driven turbine operation and maintenance strategies and reducing downtimes. We present a distributed federated machine learning approach that leaves the data on the wind turbines to preserve the data privacy, as desired by manufacturers, while still enabling fleet-wide learning on those local data. We demonstrate in a case study that wind turbines which are scarce in representative training data benefit from more accurate fault detection models with federated learning, while no turbine experiences a loss in model performance by participating in the federated learning process. When comparing conventional and federated training processes, the average model training time rises significantly by a factor of 7 in the federated training due to increased communication and overhead operations. Thus, model training times might constitute an impediment that needs to be further explored and alleviated in federated learning applications, especially for large wind turbine fleets.

Generating new fonts is a time-consuming and labor-intensive, especially in a language with a huge amount of characters like Chinese. Various deep learning models have demonstrated the ability to efficiently generate new fonts with a few reference characters of that style. This project aims to develop a few-shot cross-lingual font generator based on AGIS-Net and improve the performance metrics mentioned. Our approaches include redesigning the encoder and the loss function. We will validate our method on multiple languages and datasets mentioned.

We present ObjectMatch, a semantic and object-centric camera pose estimation for RGB-D SLAM pipelines. Modern camera pose estimators rely on direct correspondences of overlapping regions between frames; however, they cannot align camera frames with little or no overlap. In this work, we propose to leverage indirect correspondences obtained via semantic object identification. For instance, when an object is seen from the front in one frame and from the back in another frame, we can provide additional pose constraints through canonical object correspondences. We first propose a neural network to predict such correspondences on a per-pixel level, which we then combine in our energy formulation with state-of-the-art keypoint matching solved with a joint Gauss-Newton optimization. In a pairwise setting, our method improves registration recall of state-of-the-art feature matching from 77% to 87% overall and from 21% to 52% in pairs with 10% or less inter-frame overlap. In registering RGB-D sequences, our method outperforms cutting-edge SLAM baselines in challenging, low frame-rate scenarios, achieving more than 35% reduction in trajectory error in multiple scenes.