المؤلفون: Koedijk, Matthijs, Onderzoeker, Renden, P.G. (Peter), Onderzoeker, Oudejans, Raôul R.D, Onderzoeker, Kleygrewe, Lisanne, Onderzoeker, Hutter, R.I. (Vana), Onderzoeker

المساهمون: Faculteit Gezondheid, Voeding & Sport, De Haagse Hogeschool

المصدر: Frontiers in Psychology. 12((feb. 2021) (artikel 589258)):1-14

مصطلحات موضوعية: psychological competence, assessment instrument, behavioral observation, psychological observation, professional performance, police

الوصف: This paper proposes and showcases a methodology to develop an observational behavior assessment instrument to assess psychological competencies of police officers. We outline a step-by-step methodology for police organizations to measure and evaluate behavior in a meaningful way to assess these competencies. We illustrate the proposed methodology with a practical example. We posit that direct behavioral observation can be key in measuring the expression of psychological competence in practice, and that psychological competence in practice is what police organizations should care about. We hope this paper offers police organizations a methodology to perform scientifically informed observational behavior assessment of their police officers’ psychological competencies and inspires additional research efforts into this important area.

الوصول الحر: https://surfsharekit.nl/public/47340862-bb90-4753-9b13-5e52390e14daTest
https://surfsharekit.nl/objectstore/dc32ef15-c587-40ec-8438-0bfe6151d87fTest

الإتاحة: http://www.hbo-kennisbank.nl/en/page/hborecord.view/?uploadId=sharekit_hh:oai:surfsharekit.nl:47340862-bb90-4753-9b13-5e52390e14daTest

المؤلفون: Akerstrom, F, Hutter, J, Charitakis, E, Tabrizi, F, Asaad, F, Bastani, H, Bourke, T, Braunschweig, F, Drca, N, Englund, A, Friberg, L, Insulander, P, Jonsson, AH, Kenneback, G, Paul-Nordin, A, Sadigh, B, Saluveer, O, Saygi, S, Schwieler, J, Svennberg, E, Tapanainen, J, Turkmen, Y, Jensen-Urstad, M

المصدر: Heart (British Cardiac Society). 110(3):163-169

مصطلحات موضوعية: Medicin och hälsovetenskap

الوصول الحر: http://kipublications.ki.se/Default.aspx?queryparsed=id:153788335Test

المؤلفون: Everaars, L., Student, Hutter, C., Horst, G., Strikwold, M., Oudhuis, A.A.C.M., Woudstra, F.S., Hogeschool Van Hall Larenstein, Applied Research Center, Hogeschool Van Hall Larenstein

مصطلحات موضوعية: Melkvee, Dierproeven, Proefdieren

الوصف: Alternatives to animal testing are increasingly being used to assess the toxicological safety of substances and medicines, with a great deal of focus in recent years on alternatives for humans and small laboratory animals. Cattle, however, are also used for veterinary medicine research and education, but no complete alternative for testing with cattle is available so far. In analogue for rat and human, in vitro and in silico models could be developed for the cow as an alternative for animal testing. However, reliance on in vitro data entails a number of new challenges associated with translating the in vitro results to corresponding in vivo values. The RAAK-PRO project 'Generic bovine kinetic modelling platform, works on developing an in silico in vitro based model to predict the transfer of substances from feed to food (milk and meat from the cow) without testing in living cows. For this purpose, several in vitro assays for absorption are being developed. The aim of this study was to set-up a Caco-2 permeability assay to predict in vivo oral uptake values for the cow. Besides this, the application of an in vitro tiny-TIM digestion system for probiotic survival in the human GI-tract was studied, which can be used as an alternative for human studies.

الوصول الحر: https://www.greeni.nl/iguana/CMS.MetaDataEditDownload.cls?file=2:150844:1Test
https://www.greeni.nl/iguana/www.main.cls?surl=greenisearch#RecordId=2.150844Test

الإتاحة: http://www.hbo-kennisbank.nl/en/page/hborecord.view/?uploadId=samhao:oai:www.greeni.nl:VBS:2:150844Test

المؤلفون: Robertson, Jake, Hollmann, Noah, Awad, Noor, Hutter, Frank

مصطلحات موضوعية: Computer Science - Machine Learning, Computer Science - Artificial Intelligence, Computer Science - Computers and Society

الوصف: Machine Learning systems are increasingly prevalent across healthcare, law enforcement, and finance but often operate on historical data, which may carry biases against certain demographic groups. Causal and counterfactual fairness provides an intuitive way to define fairness that closely aligns with legal standards. Despite its theoretical benefits, counterfactual fairness comes with several practical limitations, largely related to the reliance on domain knowledge and approximate causal discovery techniques in constructing a causal model. In this study, we take a fresh perspective on counterfactually fair prediction, building upon recent work in in context learning (ICL) and prior fitted networks (PFNs) to learn a transformer called FairPFN. This model is pretrained using synthetic fairness data to eliminate the causal effects of protected attributes directly from observational data, removing the requirement of access to the correct causal model in practice. In our experiments, we thoroughly assess the effectiveness of FairPFN in eliminating the causal impact of protected attributes on a series of synthetic case studies and real world datasets. Our findings pave the way for a new and promising research area: transformers for causal and counterfactual fairness.

الوصول الحر: http://arxiv.org/abs/2407.05732Test

المؤلفون: Pan, Yang, Hutter, Clemens, Bölcskei, Helmut

مصطلحات موضوعية: Computer Science - Machine Learning, Computer Science - Information Theory, Mathematics - Dynamical Systems

الوصف: This paper is concerned with the fundamental limits of nonlinear dynamical system learning from input-output traces. Specifically, we show that recurrent neural networks (RNNs) are capable of learning nonlinear systems that satisfy a Lipschitz property and forget past inputs fast enough in a metric-entropy optimal manner. As the sets of sequence-to-sequence maps realized by the dynamical systems we consider are significantly more massive than function classes generally considered in deep neural network approximation theory, a refined metric-entropy characterization is needed, namely in terms of order, type, and generalized dimension. We compute these quantities for the classes of exponentially-decaying and polynomially-decaying Lipschitz fading-memory systems and show that RNNs can achieve them.

الوصول الحر: http://arxiv.org/abs/2407.01250Test

المؤلفون: Mower, Christopher E., Wan, Yuhui, Yu, Hongzhan, Grosnit, Antoine, Gonzalez-Billandon, Jonas, Zimmer, Matthieu, Wang, Jinlong, Zhang, Xinyu, Zhao, Yao, Zhai, Anbang, Liu, Puze, Palenicek, Daniel, Tateo, Davide, Cadena, Cesar, Hutter, Marco, Peters, Jan, Tian, Guangjian, Zhuang, Yuzheng, Shao, Kun, Quan, Xingyue, Hao, Jianye, Wang, Jun, Bou-Ammar, Haitham

مصطلحات موضوعية: Computer Science - Robotics, Computer Science - Artificial Intelligence

الوصف: We present a framework for intuitive robot programming by non-experts, leveraging natural language prompts and contextual information from the Robot Operating System (ROS). Our system integrates large language models (LLMs), enabling non-experts to articulate task requirements to the system through a chat interface. Key features of the framework include: integration of ROS with an AI agent connected to a plethora of open-source and commercial LLMs, automatic extraction of a behavior from the LLM output and execution of ROS actions/services, support for three behavior modes (sequence, behavior tree, state machine), imitation learning for adding new robot actions to the library of possible actions, and LLM reflection via human and environment feedback. Extensive experiments validate the framework, showcasing robustness, scalability, and versatility in diverse scenarios, including long-horizon tasks, tabletop rearrangements, and remote supervisory control. To facilitate the adoption of our framework and support the reproduction of our results, we have made our code open-source. You can access it at: https://github.com/huawei-noah/HEBO/tree/master/ROSLLMTest.
Comment: This document contains 26 pages and 13 figures

الوصول الحر: http://arxiv.org/abs/2406.19741Test

المؤلفون: Blauth, Simon, Bürger, Tobias, Häringer, Zacharias, Franke, Jörg, Hutter, Frank

مصطلحات موضوعية: Computer Science - Machine Learning, Computer Science - Artificial Intelligence

الوصف: In this paper, we present the Fast Optimizer Benchmark (FOB), a tool designed for evaluating deep learning optimizers during their development. The benchmark supports tasks from multiple domains such as computer vision, natural language processing, and graph learning. The focus is on convenient usage, featuring human-readable YAML configurations, SLURM integration, and plotting utilities. FOB can be used together with existing hyperparameter optimization (HPO) tools as it handles training and resuming of runs. The modular design enables integration into custom pipelines, using it simply as a collection of tasks. We showcase an optimizer comparison as a usage example of our tool. FOB can be found on GitHub: https://github.com/automl/FOBTest.
Comment: 5 pages + 12 appendix pages, submitted to AutoML Conf 2024 Workshop Track

الوصول الحر: http://arxiv.org/abs/2406.18701Test

المؤلفون: Richter, Julia, Kolvenbach, Hendrik, Valsecchi, Giorgio, Hutter, Marco

مصطلحات موضوعية: Computer Science - Robotics

الوصف: In unstructured environments the best path is not always the shortest, but needs to consider various objectives like energy efficiency, risk of failure or scientific outcome. This paper proposes a global planner, based on the A* algorithm, capable of individually considering multiple layers of map data for different cost objectives. We introduce weights between the objectives, which can be adapted to achieve a variety of optimal paths. In order to find the best of these paths, a tool for statistical path analysis is presented. Our planner was tested on exemplary lunar topographies to propose two trajectories for exploring the Aristarchus Plateau. The optimized paths significantly reduce the risk of failure while yielding more scientific value compared to a manually planned paths in the same area. The planner and analysis tool are made open-source in order to simplify mission planning for planetary scientists.
Comment: 8 pages, 19 figures, IEEE conference iSpaRo 2024

الوصول الحر: http://arxiv.org/abs/2406.16376Test

المؤلفون: Vutukur, Shishir Reddy, Brock, Heike, Busam, Benjamin, Birdal, Tolga, Hutter, Andreas, Ilic, Slobodan

المصدر: 3DV 2024

مصطلحات موضوعية: Computer Science - Computer Vision and Pattern Recognition

الوصف: Object Pose Estimation is a crucial component in robotic grasping and augmented reality. Learning based approaches typically require training data from a highly accurate CAD model or labeled training data acquired using a complex setup. We address this by learning to estimate pose from weakly labeled data without a known CAD model. We propose to use a NeRF to learn object shape implicitly which is later used to learn view-invariant features in conjunction with CNN using a contrastive loss. While NeRF helps in learning features that are view-consistent, CNN ensures that the learned features respect symmetry. During inference, CNN is used to predict view-invariant features which can be used to establish correspondences with the implicit 3d model in NeRF. The correspondences are then used to estimate the pose in the reference frame of NeRF. Our approach can also handle symmetric objects unlike other approaches using a similar training setup. Specifically, we learn viewpoint invariant, discriminative features using NeRF which are later used for pose estimation. We evaluated our approach on LM, LM-Occlusion, and T-Less dataset and achieved benchmark accuracy despite using weakly labeled data.
Comment: 3DV 2024

الوصول الحر: http://arxiv.org/abs/2406.13796Test

المؤلفون: Lindauer, Marius, Karl, Florian, Klier, Anne, Moosbauer, Julia, Tornede, Alexander, Mueller, Andreas, Hutter, Frank, Feurer, Matthias, Bischl, Bernd

مصطلحات موضوعية: Computer Science - Machine Learning

الوصف: Automated machine learning (AutoML) was formed around the fundamental objectives of automatically and efficiently configuring machine learning (ML) workflows, aiding the research of new ML algorithms, and contributing to the democratization of ML by making it accessible to a broader audience. Over the past decade, commendable achievements in AutoML have primarily focused on optimizing predictive performance. This focused progress, while substantial, raises questions about how well AutoML has met its broader, original goals. In this position paper, we argue that a key to unlocking AutoML's full potential lies in addressing the currently underexplored aspect of user interaction with AutoML systems, including their diverse roles, expectations, and expertise. We envision a more human-centered approach in future AutoML research, promoting the collaborative design of ML systems that tightly integrates the complementary strengths of human expertise and AutoML methodologies.

الوصول الحر: http://arxiv.org/abs/2406.03348Test