Refine
H-BRS Bibliography
- yes (242) (remove)
Departments, institutes and facilities
- Fachbereich Informatik (68)
- Fachbereich Angewandte Naturwissenschaften (59)
- Institut für Technik, Ressourcenschonung und Energieeffizienz (TREE) (43)
- Fachbereich Ingenieurwissenschaften und Kommunikation (38)
- Fachbereich Wirtschaftswissenschaften (37)
- Institut für funktionale Gen-Analytik (IFGA) (29)
- Fachbereich Sozialpolitik und Soziale Sicherung (22)
- Internationales Zentrum für Nachhaltige Entwicklung (IZNE) (20)
- Institut für Cyber Security & Privacy (ICSP) (16)
- Institut für Verbraucherinformatik (IVI) (16)
Document Type
- Article (101)
- Conference Object (72)
- Part of a Book (27)
- Preprint (15)
- Book (monograph, edited volume) (7)
- Research Data (6)
- Doctoral Thesis (5)
- Working Paper (4)
- Conference Proceedings (2)
- Part of Periodical (1)
Year of publication
- 2021 (242) (remove)
Language
- English (242) (remove)
Keywords
- Augmented Reality (4)
- Machine Learning (4)
- Big Data Analysis (3)
- Usable Security (3)
- recovery (3)
- AML (2)
- Authentication features (2)
- Automatic Differentiation (2)
- C sequestration (2)
- C source (2)
It has been well proved that deep networks are efficient at extracting features from a given (source) labeled dataset. However, it is not always the case that they can generalize well to other (target) datasets which very often have a different underlying distribution. In this report, we evaluate four different domain adaptation techniques for image classification tasks: DeepCORAL, DeepDomainConfusion, CDAN and CDAN+E. These techniques are unsupervised given that the target dataset dopes not carry any labels during training phase. We evaluate model performance on the office-31 dataset. A link to the github repository of this report can be found here: https://github.com/agrija9/Deep-Unsupervised-Domain-Adaptation.
Ice accumulation in the blades of wind turbines can cause them to describe anomalous rotations or no rotations at all, thus affecting the generation of electricity and power output. In this work, we investigate the problem of ice accumulation in wind turbines by framing it as anomaly detection of multi-variate time series. Our approach focuses on two main parts: first, learning low-dimensional representations of time series using a Variational Recurrent Autoencoder (VRAE), and second, using unsupervised clustering algorithms to classify the learned representations as normal (no ice accumulated) or abnormal (ice accumulated). We have evaluated our approach on a custom wind turbine time series dataset, for the two-classes problem (one normal versus one abnormal class), we obtained a classification accuracy of up to 96$\%$ on test data. For the multiple-class problem (one normal versus multiple abnormal classes), we present a qualitative analysis of the low-dimensional learned latent space, providing insights into the capacities of our approach to tackle such problem. The code to reproduce this work can be found here https://github.com/agrija9/Wind-Turbines-VRAE-Paper.
Machine learning and neural networks are now ubiquitous in sonar perception, but it lags behind the computer vision field due to the lack of data and pre-trained models specifically for sonar images. In this paper we present the Marine Debris Turntable dataset and produce pre-trained neural networks trained on this dataset, meant to fill the gap of missing pre-trained models for sonar images. We train Resnet 20, MobileNets, DenseNet121, SqueezeNet, MiniXception, and an Autoencoder, over several input image sizes, from 32 x 32 to 96 x 96, on the Marine Debris turntable dataset. We evaluate these models using transfer learning for low-shot classification in the Marine Debris Watertank and another dataset captured using a Gemini 720i sonar. Our results show that in both datasets the pre-trained models produce good features that allow good classification accuracy with low samples (10-30 samples per class). The Gemini dataset validates that the features transfer to other kinds of sonar sensors. We expect that the community benefits from the public release of our pre-trained models and the turntable dataset.
Object-Based Trace Model for Automatic Indicator Computation in the Human Learning Environments
(2021)
This paper proposes a traces model in the form of an object or class model (in the UML sense) which allows the automatic calculation of indicators of various kinds and independently of the computer environment for human learning (CEHL). The model is based on the establishment of a trace-based system that encompasses all the logic of traces collecting and indicators calculation. It is im-plemented in the form of a trace database. It is an important contribution in the field of the exploitation of the traces of apprenticeship in a CEHL because it pro-vides a general formalism for modeling the traces and allowing the calculation of several indicators at the same time. Also, with the inclusion of calculated indica-tors as potential learning traces, our model provides a formalism for classifying the various indicators in the form of inheritance relationships, which promotes the reuse of indicators already calculated. Economically, the model can allow organi-zations with different learning platforms to invest only in one traces Management System. At the social level, it can allow a better sharing of trace databases be-tween the various research institutions in the field of CEHL.
The universal basic income grant (UBIG): A comparative review of the characteristics and impact
(2021)
In recent years, public debates, pilot projects and academic research have internationally boosted the prominence of the universal basic income grant (UBIG) as a policy option. Despite this prominence, the arguments and evidence of the UBIG discussion have not been systematically put forward and discussed in light of the different UBIG conceptual understandings and applications. This paper adds value to the debate by systematic presenting the social, economical and political arguments in support of and against a UBIG. It furthermore discusses the UBIG dimensions/characteristics and variations, and also pose questions about whether all the UBIG experiments can really be classified as a UBIG. Antagonist of a UBIG often raise concerns about the negative effect of the lack of conditions and targeting in a UBIG. Since evidence on the impact of UBIG is limited, this paper turns to the evidence base on unconditional cash transfers and conditional cash transfers. The results show that it is the cash transfer rather than the conditionality and targeting that produce positive outcomes in areas of personal wellbeing.
This study investigated the application potential of Black Soldier Fly Larva Hermetia illucens Stratiomyidae: Diptera (L.1758) for wastewater treatment and the removal potential of chemical oxygen demand, ammonia, and phosphorus of and liquid manure residue and municipal waste water containing 1% solids content. Black Soldier Fly Larva were found to reduce the concentration of chemical oxygen demand, but unfortunately, increase the concentration of ammonia and phosphorus. The ability of Black Soldier Fly Larva to feed on organic waste of Liquid manure residue showed that Black Soldier Fly Larva increase their weight by 365% in a solution with 12% solids content and by 595% in a solution having 6% solids content. The study also showed that Black Soldier Fly Larva have the ability to survive in a solution of 1% solids content and have the ability to reduce chemical oxygen demand by up to 86.4% for liquid manure residue and 46.9% for municipal wastewater after 24 hours. Generally, ammonia increased by 43.9% for Liquid manure residue and 98.6% for municipal wastewater. Total phosphorus showed an increase of 11.0% and 88.6% increase for liquid manure residue and municipal wastewater respectively over the 8-day study. Transparent environments tend to reduce the COD content more than the dark environment, both for the liquid manure residue (55.8% and 65.4%) and municipal wastewater (71.5% and 66.4%).
The idea of a basic income grant (BIG) is not new and there are ongoing debates internationally as well as nationally in low- and middle-income countries just like in high-income countries of a BIG as a social protection policy option. The challenge is that there are different conceptualisations, which conflates and muddles the understanding. In the context of social assistance provision, a universal basic income grant (UBIG) is often compared and contrasted against targeted cash transfers (CTs). This case study systematically presents the arguments for targeted CTs and UBIGs. The value of the case study is that it systematically brings together these arguments, highlighting the variations in UBIG applications, including the evidence and actual impact of UBIG experiments. The structure of the case study is as follows: Section 2 simultaneously contrasts and compares the arguments for targeted CTs and UBIG. Section 3 discusses UBIG experiments, as well as presenting the evidence on the application of the UBIG idea, and Section 4 concludes.
Robot Action Diagnosis and Experience Correction by Falsifying Parameterised Execution Models
(2021)
When faced with an execution failure, an intelligent robot should be able to identify the likely reasons for the failure and adapt its execution policy accordingly. This paper addresses the question of how to utilise knowledge about the execution process, expressed in terms of learned constraints, in order to direct the diagnosis and experience acquisition process. In particular, we present two methods for creating a synergy between failure diagnosis and execution model learning. We first propose a method for diagnosing execution failures of parameterised action execution models, which searches for action parameters that violate a learned precondition model. We then develop a strategy that uses the results of the diagnosis process for generating synthetic data that are more likely to lead to successful execution, thereby increasing the set of available experiences to learn from. The diagnosis and experience correction methods are evaluated for the problem of handle grasping, such that we experimentally demonstrate the effectiveness of the diagnosis algorithm and show that corrected failed experiences can contribute towards improving the execution success of a robot.
The majority of biomedical knowledge is stored in structured databases or as unstructured text in scientific publications. This vast amount of information has led to numerous machine learning-based biological applications using either text through natural language processing (NLP) or structured data through knowledge graph embedding models (KGEMs). However, representations based on a single modality are inherently limited. To generate better representations of biological knowledge, we propose STonKGs, a Sophisticated Transformer trained on biomedical text and Knowledge Graphs. This multimodal Transformer uses combined input sequences of structured information from KGs and unstructured text data from biomedical literature to learn joint representations. First, we pre-trained STonKGs on a knowledge base assembled by the Integrated Network and Dynamical Reasoning Assembler (INDRA) consisting of millions of text-triple pairs extracted from biomedical literature by multiple NLP systems. Then, we benchmarked STonKGs against two baseline models trained on either one of the modalities (i.e., text or KG) across eight different classification tasks, each corresponding to a different biological application. Our results demonstrate that STonKGs outperforms both baselines, especially on the more challenging tasks with respect to the number of classes, improving upon the F1-score of the best baseline by up to 0.083. Additionally, our pre-trained model as well as the model architecture can be adapted to various other transfer learning applications. Finally, the source code and pre-trained STonKGs models are available at https://github.com/stonkgs/stonkgs and https://huggingface.co/stonkgs/stonkgs-150k.
A qualitative study of Machine Learning practices and engineering challenges in Earth Observation
(2021)
Machine Learning (ML) is ubiquitously on the advance. Like many domains, Earth Observation (EO) also increasingly relies on ML applications, where ML methods are applied to process vast amounts of heterogeneous and continuous data streams to answer socially and environmentally relevant questions. However, developing such ML- based EO systems remains challenging: Development processes and employed workflows are often barely structured and poorly reported. The application of ML methods and techniques is considered to be opaque and the lack of transparency is contradictory to the responsible development of ML-based EO applications. To improve this situation a better understanding of the current practices and engineering-related challenges in developing ML-based EO applications is required. In this paper, we report observations from an exploratory study where five experts shared their view on ML engineering in semi-structured interviews. We analysed these interviews with coding techniques as often applied in the domain of empirical software engineering. The interviews provide informative insights into the practical development of ML applications and reveal several engineering challenges. In addition, interviewees participated in a novel workflow sketching task, which provided a tangible reflection of implicit processes. Overall, the results confirm a gap between theoretical conceptions and real practices in ML development even though workflows were sketched abstractly as textbook-like. The results pave the way for a large-scale investigation on requirements for ML engineering in EO.
When an autonomous robot learns how to execute actions, it is of interest to know if and when the execution policy can be generalised to variations of the learning scenarios. This can inform the robot about the necessity of additional learning, as using incomplete or unsuitable policies can lead to execution failures. Generalisation is particularly relevant when a robot has to deal with a large variety of objects and in different contexts. In this paper, we propose and analyse a strategy for generalising parameterised execution models of manipulation actions over different objects based on an object ontology. In particular, a robot transfers a known execution model to objects of related classes according to the ontology, but only if there is no other evidence that the model may be unsuitable. This allows using ontological knowledge as prior information that is then refined by the robot’s own experiences. We verify our algorithm for two actions – grasping and stowing everyday objects – such that we show that the robot can deduce cases in which an existing policy can generalise to other objects and when additional execution knowledge has to be acquired.
Target meaning representations for semantic parsing tasks are often based on programming or query languages, such as SQL, and can be formalized by a context-free grammar. Assuming a priori knowledge of the target domain, such grammars can be exploited to enforce syntactical constraints when predicting logical forms. To that end, we assess how syntactical parsers can be integrated into modern encoder-decoder frameworks. Specifically, we implement an attentional SEQ2SEQ model that uses an LR parser to maintain syntactically valid sequences throughout the decoding procedure. Compared to other approaches to grammar-guided decoding that modify the underlying neural network architecture or attempt to derive full parse trees, our approach is conceptually simpler, adds less computational overhead during inference and integrates seamlessly with current SEQ2SEQ frameworks. We present preliminary evaluation results against a recurrent SEQ2SEQ baseline on GEOQUERY and ATIS and demonstrate improved performance while enforcing grammatical constraints.