@article{CeulemansGuillGaedke2021, author = {Ceulemans, Ruben and Guill, Christian and Gaedke, Ursula}, title = {Top predators govern multitrophic diversity effects in tritrophic food webs}, series = {Ecology : a publication of the Ecological Society of America}, volume = {102}, journal = {Ecology : a publication of the Ecological Society of America}, number = {7}, publisher = {Wiley}, address = {Hoboken}, issn = {0012-9658}, doi = {10.1002/ecy.3379}, pages = {16}, year = {2021}, abstract = {It is well known that functional diversity strongly affects ecosystem functioning. However, even in rather simple model communities consisting of only two or, at best, three trophic levels, the relationship between multitrophic functional diversity and ecosystem functioning appears difficult to generalize, because of its high contextuality. In this study, we considered several differently structured tritrophic food webs, in which the amount of functional diversity was varied independently on each trophic level. To achieve generalizable results, largely independent of parametrization, we examined the outcomes of 128,000 parameter combinations sampled from ecologically plausible intervals, with each tested for 200 randomly sampled initial conditions. Analysis of our data was done by training a random forest model. This method enables the identification of complex patterns in the data through partial dependence graphs, and the comparison of the relative influence of model parameters, including the degree of diversity, on food-web properties. We found that bottom-up and top-down effects cascade simultaneously throughout the food web, intimately linking the effects of functional diversity of any trophic level to the amount of diversity of other trophic levels, which may explain the difficulty in unifying results from previous studies. Strikingly, only with high diversity throughout the whole food web, different interactions synergize to ensure efficient exploitation of the available nutrients and efficient biomass transfer to higher trophic levels, ultimately leading to a high biomass and production on the top level. The temporal variation of biomass showed a more complex pattern with increasing multitrophic diversity: while the system initially became less variable, eventually the temporal variation rose again because of the increasingly complex dynamical patterns. Importantly, top predator diversity and food-web parameters affecting the top trophic level were of highest importance to determine the biomass and temporal variability of any trophic level. Overall, our study reveals that the mechanisms by which diversity influences ecosystem functioning are affected by every part of the food web, hampering the extrapolation of insights from simple monotrophic or bitrophic systems to complex natural food webs.}, language = {en} } @article{SteinbergVasyuraBathkeGaebleretal.2021, author = {Steinberg, Andreas and Vasyura-Bathke, Hannes and Gaebler, Peter Jost and Ohrnberger, Matthias and Ceranna, Lars}, title = {Estimation of seismic moment tensors using variational inference machine learning}, series = {Journal of geophysical research : Solid earth}, volume = {126}, journal = {Journal of geophysical research : Solid earth}, number = {10}, publisher = {American Geophysical Union}, address = {Washington}, issn = {2169-9313}, doi = {10.1029/2021JB022685}, pages = {16}, year = {2021}, abstract = {We present an approach for rapidly estimating full moment tensors of earthquakes and their parameter uncertainties based on short time windows of recorded seismic waveform data by considering deep learning of Bayesian Neural Networks (BNNs). The individual neural networks are trained on synthetic seismic waveform data and corresponding known earthquake moment-tensor parameters. A monitoring volume has been predefined to form a three-dimensional grid of locations and to train a BNN for each grid point. Variational inference on several of these networks allows us to consider several sources of error and how they affect the estimated full moment-tensor parameters and their uncertainties. In particular, we demonstrate how estimated parameter distributions are affected by uncertainties in the earthquake centroid location in space and time as well as in the assumed Earth structure model. We apply our approach as a proof of concept on seismic waveform recordings of aftershocks of the Ridgecrest 2019 earthquake with moment magnitudes ranging from Mw 2.7 to Mw 5.5. Overall, good agreement has been achieved between inferred parameter ensembles and independently estimated parameters using classical methods. Our developed approach is fast and robust, and therefore, suitable for down-stream analyses that need rapid estimates of the source mechanism for a large number of earthquakes.}, language = {en} } @article{CopeBaukmannKlingeretal.2021, author = {Cope, Justin L. and Baukmann, Hannes A. and Klinger, J{\"o}rn E. and Ravarani, Charles N. J. and B{\"o}ttinger, Erwin and Konigorski, Stefan and Schmidt, Marco F.}, title = {Interaction-based feature selection algorithm outperforms polygenic risk score in predicting Parkinson's Disease status}, series = {Frontiers in genetics}, volume = {12}, journal = {Frontiers in genetics}, publisher = {Frontiers Media}, address = {Lausanne}, issn = {1664-8021}, doi = {10.3389/fgene.2021.744557}, pages = {9}, year = {2021}, abstract = {Polygenic risk scores (PRS) aggregating results from genome-wide association studies are the state of the art in the prediction of susceptibility to complex traits or diseases, yet their predictive performance is limited for various reasons, not least of which is their failure to incorporate the effects of gene-gene interactions. Novel machine learning algorithms that use large amounts of data promise to find gene-gene interactions in order to build models with better predictive performance than PRS. Here, we present a data preprocessing step by using data-mining of contextual information to reduce the number of features, enabling machine learning algorithms to identify gene-gene interactions. We applied our approach to the Parkinson's Progression Markers Initiative (PPMI) dataset, an observational clinical study of 471 genotyped subjects (368 cases and 152 controls). With an AUC of 0.85 (95\% CI = [0.72; 0.96]), the interaction-based prediction model outperforms the PRS (AUC of 0.58 (95\% CI = [0.42; 0.81])). Furthermore, feature importance analysis of the model provided insights into the mechanism of Parkinson's disease. For instance, the model revealed an interaction of previously described drug target candidate genes TMEM175 and GAPDHP25. These results demonstrate that interaction-based machine learning models can improve genetic prediction models and might provide an answer to the missing heritability problem.}, language = {en} } @misc{KonakWegnerArnrich2021, author = {Konak, Orhan and Wegner, Pit and Arnrich, Bert}, title = {IMU-Based Movement Trajectory Heatmaps for Human Activity Recognition}, series = {Postprints der Universit{\"a}t Potsdam : Reihe der Digital Engineering Fakult{\"a}t}, journal = {Postprints der Universit{\"a}t Potsdam : Reihe der Digital Engineering Fakult{\"a}t}, number = {4}, doi = {10.25932/publishup-48779}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-487799}, pages = {17}, year = {2021}, abstract = {Recent trends in ubiquitous computing have led to a proliferation of studies that focus on human activity recognition (HAR) utilizing inertial sensor data that consist of acceleration, orientation and angular velocity. However, the performances of such approaches are limited by the amount of annotated training data, especially in fields where annotating data is highly time-consuming and requires specialized professionals, such as in healthcare. In image classification, this limitation has been mitigated by powerful oversampling techniques such as data augmentation. Using this technique, this work evaluates to what extent transforming inertial sensor data into movement trajectories and into 2D heatmap images can be advantageous for HAR when data are scarce. A convolutional long short-term memory (ConvLSTM) network that incorporates spatiotemporal correlations was used to classify the heatmap images. Evaluation was carried out on Deep Inertial Poser (DIP), a known dataset composed of inertial sensor data. The results obtained suggest that for datasets with large numbers of subjects, using state-of-the-art methods remains the best alternative. However, a performance advantage was achieved for small datasets, which is usually the case in healthcare. Moreover, movement trajectories provide a visual representation of human activities, which can help researchers to better interpret and analyze motion patterns.}, language = {en} } @phdthesis{Lazaridou2021, author = {Lazaridou, Konstantina}, title = {Revealing hidden patterns in political news and social media with machine learning}, doi = {10.25932/publishup-50273}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-502734}, school = {Universit{\"a}t Potsdam}, pages = {xv, 140}, year = {2021}, abstract = {As part of our everyday life we consume breaking news and interpret it based on our own viewpoints and beliefs. We have easy access to online social networking platforms and news media websites, where we inform ourselves about current affairs and often post about our own views, such as in news comments or social media posts. The media ecosystem enables opinions and facts to travel from news sources to news readers, from news article commenters to other readers, from social network users to their followers, etc. The views of the world many of us have depend on the information we receive via online news and social media. Hence, it is essential to maintain accurate, reliable and objective online content to ensure democracy and verity on the Web. To this end, we contribute to a trustworthy media ecosystem by analyzing news and social media in the context of politics to ensure that media serves the public interest. In this thesis, we use text mining, natural language processing and machine learning techniques to reveal underlying patterns in political news articles and political discourse in social networks. Mainstream news sources typically cover a great amount of the same news stories every day, but they often place them in a different context or report them from different perspectives. In this thesis, we are interested in how distinct and predictable newspaper journalists are, in the way they report the news, as a means to understand and identify their different political beliefs. To this end, we propose two models that classify text from news articles to their respective original news source, i.e., reported speech and also news comments. Our goal is to capture systematic quoting and commenting patterns by journalists and news commenters respectively, which can lead us to the newspaper where the quotes and comments are originally published. Predicting news sources can help us understand the potential subjective nature behind news storytelling and the magnitude of this phenomenon. Revealing this hidden knowledge can restore our trust in media by advancing transparency and diversity in the news. Media bias can be expressed in various subtle ways in the text and it is often challenging to identify these bias manifestations correctly, even for humans. However, media experts, e.g., journalists, are a powerful resource that can help us overcome the vague definition of political media bias and they can also assist automatic learners to find the hidden bias in the text. Due to the enormous technological advances in artificial intelligence, we hypothesize that identifying political bias in the news could be achieved through the combination of sophisticated deep learning modelsxi and domain expertise. Therefore, our second contribution is a high-quality and reliable news dataset annotated by journalists for political bias and a state-of-the-art solution for this task based on curriculum learning. Our aim is to discover whether domain expertise is necessary for this task and to provide an automatic solution for this traditionally manually-solved problem. User generated content is fundamentally different from news articles, e.g., messages are shorter, they are often personal and opinionated, they refer to specific topics and persons, etc. Regarding political and socio-economic news, individuals in online communities make use of social networks to keep their peers up-to-date and to share their own views on ongoing affairs. We believe that social media is also an as powerful instrument for information flow as the news sources are, and we use its unique characteristic of rapid news coverage for two applications. We analyze Twitter messages and debate transcripts during live political presidential debates to automatically predict the topics that Twitter users discuss. Our goal is to discover the favoured topics in online communities on the dates of political events as a way to understand the political subjects of public interest. With the up-to-dateness of microblogs, an additional opportunity emerges, namely to use social media posts and leverage the real-time verity about discussed individuals to find their locations. That is, given a person of interest that is mentioned in online discussions, we use the wisdom of the crowd to automatically track her physical locations over time. We evaluate our approach in the context of politics, i.e., we predict the locations of US politicians as a proof of concept for important use cases, such as to track people that are national risks, e.g., warlords and wanted criminals.}, language = {en} } @phdthesis{Loster2021, author = {Loster, Michael}, title = {Knowledge base construction with machine learning methods}, doi = {10.25932/publishup-50145}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-501459}, school = {Universit{\"a}t Potsdam}, pages = {ii, 130}, year = {2021}, abstract = {Modern knowledge bases contain and organize knowledge from many different topic areas. Apart from specific entity information, they also store information about their relationships amongst each other. Combining this information results in a knowledge graph that can be particularly helpful in cases where relationships are of central importance. Among other applications, modern risk assessment in the financial sector can benefit from the inherent network structure of such knowledge graphs by assessing the consequences and risks of certain events, such as corporate insolvencies or fraudulent behavior, based on the underlying network structure. As public knowledge bases often do not contain the necessary information for the analysis of such scenarios, the need arises to create and maintain dedicated domain-specific knowledge bases. This thesis investigates the process of creating domain-specific knowledge bases from structured and unstructured data sources. In particular, it addresses the topics of named entity recognition (NER), duplicate detection, and knowledge validation, which represent essential steps in the construction of knowledge bases. As such, we present a novel method for duplicate detection based on a Siamese neural network that is able to learn a dataset-specific similarity measure which is used to identify duplicates. Using the specialized network architecture, we design and implement a knowledge transfer between two deduplication networks, which leads to significant performance improvements and a reduction of required training data. Furthermore, we propose a named entity recognition approach that is able to identify company names by integrating external knowledge in the form of dictionaries into the training process of a conditional random field classifier. In this context, we study the effects of different dictionaries on the performance of the NER classifier. We show that both the inclusion of domain knowledge as well as the generation and use of alias names results in significant performance improvements. For the validation of knowledge represented in a knowledge base, we introduce Colt, a framework for knowledge validation based on the interactive quality assessment of logical rules. In its most expressive implementation, we combine Gaussian processes with neural networks to create Colt-GP, an interactive algorithm for learning rule models. Unlike other approaches, Colt-GP uses knowledge graph embeddings and user feedback to cope with data quality issues of knowledge bases. The learned rule model can be used to conditionally apply a rule and assess its quality. Finally, we present CurEx, a prototypical system for building domain-specific knowledge bases from structured and unstructured data sources. Its modular design is based on scalable technologies, which, in addition to processing large datasets, ensures that the modules can be easily exchanged or extended. CurEx offers multiple user interfaces, each tailored to the individual needs of a specific user group and is fully compatible with the Colt framework, which can be used as part of the system. We conduct a wide range of experiments with different datasets to determine the strengths and weaknesses of the proposed methods. To ensure the validity of our results, we compare the proposed methods with competing approaches.}, language = {en} } @article{Ayzel2021, author = {Ayzel, Georgy}, title = {Deep neural networks in hydrology}, series = {Vestnik of Saint Petersburg University. Earth Sciences}, volume = {66}, journal = {Vestnik of Saint Petersburg University. Earth Sciences}, number = {1}, publisher = {Univ. Press}, address = {St. Petersburg}, issn = {2541-9668}, doi = {10.21638/spbu07.2021.101}, pages = {5 -- 18}, year = {2021}, abstract = {For around a decade, deep learning - the sub-field of machine learning that refers to artificial neural networks comprised of many computational layers - modifies the landscape of statistical model development in many research areas, such as image classification, machine translation, and speech recognition. Geoscientific disciplines in general and the field of hydrology in particular, also do not stand aside from this movement. Recently, the proliferation of modern deep learning-based techniques and methods has been actively gaining popularity for solving a wide range of hydrological problems: modeling and forecasting of river runoff, hydrological model parameters regionalization, assessment of available water resources. identification of the main drivers of the recent change in water balance components. This growing popularity of deep neural networks is primarily due to their high universality and efficiency. The presented qualities, together with the rapidly growing amount of accumulated environmental information, as well as increasing availability of computing facilities and resources, allow us to speak about deep neural networks as a new generation of mathematical models designed to, if not to replace existing solutions, but significantly enrich the field of geophysical processes modeling. This paper provides a brief overview of the current state of the field of development and application of deep neural networks in hydrology. Also in the following study, the qualitative long-term forecast regarding the development of deep learning technology for managing the corresponding hydrological modeling challenges is provided based on the use of "Gartner Hype Curve", which in the general details describes a life cycle of modern technologies.}, language = {en} } @article{BrandesSicksBerger2021, author = {Brandes, Stefanie and Sicks, Florian and Berger, Anne}, title = {Behaviour classification on giraffes (Giraffa camelopardalis) using machine learning algorithms on triaxial acceleration data of two commonly used GPS devices and its possible application for their management and conservation}, series = {Sensors}, volume = {21}, journal = {Sensors}, number = {6}, publisher = {MDPI}, address = {Basel}, issn = {1424-8220}, doi = {10.3390/s21062229}, pages = {22}, year = {2021}, abstract = {Averting today's loss of biodiversity and ecosystem services can be achieved through conservation efforts, especially of keystone species. Giraffes (Giraffa camelopardalis) play an important role in sustaining Africa's ecosystems, but are 'vulnerable' according to the IUCN Red List since 2016. Monitoring an animal's behavior in the wild helps to develop and assess their conservation management. One mechanism for remote tracking of wildlife behavior is to attach accelerometers to animals to record their body movement. We tested two different commercially available high-resolution accelerometers, e-obs and Africa Wildlife Tracking (AWT), attached to the top of the heads of three captive giraffes and analyzed the accuracy of automatic behavior classifications, focused on the Random Forests algorithm. For both accelerometers, behaviors of lower variety in head and neck movements could be better predicted (i.e., feeding above eye level, mean prediction accuracy e-obs/AWT: 97.6\%/99.7\%; drinking: 96.7\%/97.0\%) than those with a higher variety of body postures (such as standing: 90.7-91.0\%/75.2-76.7\%; rumination: 89.6-91.6\%/53.5-86.5\%). Nonetheless both devices come with limitations and especially the AWT needs technological adaptations before applying it on animals in the wild. Nevertheless, looking at the prediction results, both are promising accelerometers for behavioral classification of giraffes. Therefore, these devices when applied to free-ranging animals, in combination with GPS tracking, can contribute greatly to the conservation of giraffes.}, language = {en} } @article{HampfNendelStreyetal.2021, author = {Hampf, Anna and Nendel, Claas and Strey, Simone and Strey, Robert}, title = {Biotic yield losses in the Southern Amazon, Brazil}, series = {Frontiers in plant science : FPLS}, volume = {12}, journal = {Frontiers in plant science : FPLS}, publisher = {Frontiers Media}, address = {Lausanne}, issn = {1664-462X}, doi = {10.3389/fpls.2021.621168}, pages = {16}, year = {2021}, abstract = {Pathogens and animal pests (P\&A) are a major threat to global food security as they directly affect the quantity and quality of food. The Southern Amazon, Brazil's largest domestic region for soybean, maize and cotton production, is particularly vulnerable to the outbreak of P\&A due to its (sub)tropical climate and intensive farming systems. However, little is known about the spatial distribution of P\&A and the related yield losses. Machine learning approaches for the automated recognition of plant diseases can help to overcome this research gap. The main objectives of this study are to (1) evaluate the performance of Convolutional Neural Networks (ConvNets) in classifying P\&A, (2) map the spatial distribution of P\&A in the Southern Amazon, and (3) quantify perceived yield and economic losses for the main soybean and maize P\&A. The objectives were addressed by making use of data collected with the smartphone application Plantix. The core of the app's functioning is the automated recognition of plant diseases via ConvNets. Data on expected yield losses were gathered through a short survey included in an "expert" version of the application, which was distributed among agronomists. Between 2016 and 2020, Plantix users collected approximately 78,000 georeferenced P\&A images in the Southern Amazon. The study results indicate a high performance of the trained ConvNets in classifying 420 different crop-disease combinations. Spatial distribution maps and expert-based yield loss estimates indicate that maize rust, bacterial stalk rot and the fall armyworm are among the most severe maize P\&A, whereas soybean is mainly affected by P\&A like anthracnose, downy mildew, frogeye leaf spot, stink bugs and brown spot. Perceived soybean and maize yield losses amount to 12 and 16\%, respectively, resulting in annual yield losses of approximately 3.75 million tonnes for each crop and economic losses of US\$2 billion for both crops together. The high level of accuracy of the trained ConvNets, when paired with widespread use from following a citizen-science approach, results in a data source that will shed new light on yield loss estimates, e.g., for the analysis of yield gaps and the development of measures to minimise them.}, language = {en} } @article{VaidChanChaudharyetal.2021, author = {Vaid, Akhil and Chan, Lili and Chaudhary, Kumardeep and Jaladanki, Suraj K. and Paranjpe, Ishan and Russak, Adam J. and Kia, Arash and Timsina, Prem and Levin, Matthew A. and He, John Cijiang and B{\"o}ttinger, Erwin and Charney, Alexander W. and Fayad, Zahi A. and Coca, Steven G. and Glicksberg, Benjamin S. and Nadkarni, Girish N.}, title = {Predictive approaches for acute dialysis requirement and death in COVID-19}, series = {Clinical journal of the American Society of Nephrology : CJASN}, volume = {16}, journal = {Clinical journal of the American Society of Nephrology : CJASN}, number = {8}, publisher = {American Society of Nephrology}, address = {Washington}, organization = {MSCIC}, issn = {1555-9041}, doi = {10.2215/CJN.17311120}, pages = {1158 -- 1168}, year = {2021}, abstract = {Background and objectives AKI treated with dialysis initiation is a common complication of coronavirus disease 2019 (COVID-19) among hospitalized patients. However, dialysis supplies and personnel are often limited. Design, setting, participants, \& measurements Using data from adult patients hospitalized with COVID-19 from five hospitals from theMount Sinai Health System who were admitted between March 10 and December 26, 2020, we developed and validated several models (logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme GradientBoosting [XGBoost; with and without imputation]) for predicting treatment with dialysis or death at various time horizons (1, 3, 5, and 7 days) after hospital admission. Patients admitted to theMount Sinai Hospital were used for internal validation, whereas the other hospitals formed part of the external validation cohort. Features included demographics, comorbidities, and laboratory and vital signs within 12 hours of hospital admission. Results A total of 6093 patients (2442 in training and 3651 in external validation) were included in the final cohort. Of the different modeling approaches used, XGBoost without imputation had the highest area under the receiver operating characteristic (AUROC) curve on internal validation (range of 0.93-0.98) and area under the precisionrecall curve (AUPRC; range of 0.78-0.82) for all time points. XGBoost without imputation also had the highest test parameters on external validation (AUROC range of 0.85-0.87, and AUPRC range of 0.27-0.54) across all time windows. XGBoost without imputation outperformed all models with higher precision and recall (mean difference in AUROC of 0.04; mean difference in AUPRC of 0.15). Features of creatinine, BUN, and red cell distribution width were major drivers of the model's prediction. Conclusions An XGBoost model without imputation for prediction of a composite outcome of either death or dialysis in patients positive for COVID-19 had the best performance, as compared with standard and other machine learning models.}, language = {en} } @article{EbersHochRosenkranzetal.2021, author = {Ebers, Martin and Hoch, Veronica R. S. and Rosenkranz, Frank and Ruschemeier, Hannah and Steinr{\"o}tter, Bj{\"o}rn}, title = {The European Commission's proposal for an Artificial Intelligence Act}, series = {J : multidisciplinary scientific journal}, volume = {4}, journal = {J : multidisciplinary scientific journal}, number = {4}, publisher = {MDPI}, address = {Basel}, issn = {2571-8800}, doi = {10.3390/j4040043}, pages = {589 -- 603}, year = {2021}, abstract = {On 21 April 2021, the European Commission presented its long-awaited proposal for a Regulation "laying down harmonized rules on Artificial Intelligence", the so-called "Artificial Intelligence Act" (AIA). This article takes a critical look at the proposed regulation. After an introduction (1), the paper analyzes the unclear preemptive effect of the AIA and EU competences (2), the scope of application (3), the prohibited uses of Artificial Intelligence (AI) (4), the provisions on high-risk AI systems (5), the obligations of providers and users (6), the requirements for AI systems with limited risks (7), the enforcement system (8), the relationship of the AIA with the existing legal framework (9), and the regulatory gaps (10). The last section draws some final conclusions (11).}, language = {en} } @misc{PanzerBenderGronau2021, author = {Panzer, Marcel and Bender, Benedict and Gronau, Norbert}, title = {Deep reinforcement learning in production planning and control}, series = {Zweitver{\"o}ffentlichungen der Universit{\"a}t Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe}, journal = {Zweitver{\"o}ffentlichungen der Universit{\"a}t Potsdam : Wirtschafts- und Sozialwissenschaftliche Reihe}, issn = {2701-6277}, doi = {10.25932/publishup-60572}, url = {http://nbn-resolving.de/urn:nbn:de:kobv:517-opus4-605722}, pages = {13}, year = {2021}, abstract = {Increasingly fast development cycles and individualized products pose major challenges for today's smart production systems in times of industry 4.0. The systems must be flexible and continuously adapt to changing conditions while still guaranteeing high throughputs and robustness against external disruptions. Deep reinforcement learning (RL) algorithms, which already reached impressive success with Google DeepMind's AlphaGo, are increasingly transferred to production systems to meet related requirements. Unlike supervised and unsupervised machine learning techniques, deep RL algorithms learn based on recently collected sensorand process-data in direct interaction with the environment and are able to perform decisions in real-time. As such, deep RL algorithms seem promising given their potential to provide decision support in complex environments, as production systems, and simultaneously adapt to changing circumstances. While different use-cases for deep RL emerged, a structured overview and integration of findings on their application are missing. To address this gap, this contribution provides a systematic literature review of existing deep RL applications in the field of production planning and control as well as production logistics. From a performance perspective, it became evident that deep RL can beat heuristics significantly in their overall performance and provides superior solutions to various industrial use-cases. Nevertheless, safety and reliability concerns must be overcome before the widespread use of deep RL is possible which presumes more intensive testing of deep RL in real world applications besides the already ongoing intensive simulations.}, language = {en} } @inproceedings{PanzerBenderGronau2021, author = {Panzer, Marcel and Bender, Benedict and Gronau, Norbert}, title = {Deep reinforcement learning in production planning and control}, series = {Proceedings of the Conference on Production Systems and Logistics}, booktitle = {Proceedings of the Conference on Production Systems and Logistics}, publisher = {publish-Ing.}, address = {Hannover}, issn = {2701-6277}, doi = {10.15488/11238}, pages = {535 -- 545}, year = {2021}, abstract = {Increasingly fast development cycles and individualized products pose major challenges for today's smart production systems in times of industry 4.0. The systems must be flexible and continuously adapt to changing conditions while still guaranteeing high throughputs and robustness against external disruptions. Deep rein- forcement learning (RL) algorithms, which already reached impressive success with Google DeepMind's AlphaGo, are increasingly transferred to production systems to meet related requirements. Unlike supervised and unsupervised machine learning techniques, deep RL algorithms learn based on recently collected sensor- and process-data in direct interaction with the environment and are able to perform decisions in real-time. As such, deep RL algorithms seem promising given their potential to provide decision support in complex environments, as production systems, and simultaneously adapt to changing circumstances. While different use-cases for deep RL emerged, a structured overview and integration of findings on their application are missing. To address this gap, this contribution provides a systematic literature review of existing deep RL applications in the field of production planning and control as well as production logistics. From a performance perspective, it became evident that deep RL can beat heuristics significantly in their overall performance and provides superior solutions to various industrial use-cases. Nevertheless, safety and reliability concerns must be overcome before the widespread use of deep RL is possible which presumes more intensive testing of deep RL in real world applications besides the already ongoing intensive simulations.}, language = {en} }