检索结果-内蒙古大学图书馆

ProbSAINT: Probabilistic Tabular Regression for Used Car Pricing

学校读者我要写书评

暂无评论

ProbSAINT: Probabilistic Tabular Regression for Used Car Pri...

IEEE International Conference on Big data

作者： Kiran Madhusudhanan Gunnar Behrens Maximilian Stubbemann Lars Schmidt-Thieme Information Systems and Machine Learning Lab VWFS Data Analytics Research Center University Of Hildesheim Data Analytics & AI Engineering Volkswagen Financial Services AG

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

Used car pricing is a critical aspect of the automotive industry, influenced by many economic factors and market dynamics. With the recent surge in online marketplaces and increased demand for used cars, accurate pricing would benefit both buyers and sellers by ensuring fair transactions. However, the transition towards automated pricing algorithms using machine learning necessitates the comprehension of model uncertainties, specifically the ability to flag predictions that the model is unsure about. Although recent literature proposes the use of boosting algorithms or nearest neighbor-based approaches for swift and precise price predictions, encapsulating model uncertainties with such algorithms presents a complex challenge. We introduce ProbSAINT, a model that offers a principled approach for uncertainty quantification of its price predictions, along with accurate point predictions that are comparable to state-of-the-art boosting techniques. Furthermore, acknowledging that the business prefers pricing used cars based on the number of days the vehicle was listed for sale, we show how ProbSAINT can be used as a dynamic forecasting model for predicting price probabilities for different expected offer durations. Our experiments further indicate that ProbSAINT is especially accurate in instances where it is highly certain. This proves the applicability of its probabilistic predictions in real-world scenarios where trustworthiness is crucial.

关键词： Industries Accuracy Uncertainty machine learning algorithms Pricing Predictive models Probabilistic logic Prediction algorithms Automobiles Vehicle dynamics

Sensor-based Stride Segmentation and Gait Parameter Extraction Using a Hidden Markov Model in Patients with Hereditary Spastic Paraplegia

学校读者我要写书评

暂无评论

Sensor-based Stride Segmentation and Gait Parameter Extracti...

IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES)

作者： Alzhraa A. Ibrahim Veronika Koch Verena Jakob Arne Küderle Malte Ollenschlager Kathrin Kotter Bjoern M. Eskofier Martin Regensburger Jürgen Winkler Heiko Gaßner Department of Molecular Neurology University Hospital Erlangen Erlangen Germany Fraunhofer IIS Machine Learning and Data Analytics Lab FAU Fraunhofer Institute for Integrated Circuits Erlangen Germany Machine Learning and Data Analytics Lab FAU Erlangen Germany

ISBN: (数字)9798350383409

ISBN: (纸本)9798350383416

Hereditary Spastic Paraplegia (HSP) comprises a group of neurodegenerative disorders causing progressive lower limb symptoms primarily affecting gait functions. Progressive symmetric gait spasticity poses challenges for gait analysis. Traditional motion capture systems offer precise gait parameters but are confined to laboratory settings. In contrast, wearable sensor technology allows for gait analysis settings in real-world environments. Yet, accurate segmentation of gait cycles and ex-traction of temporal parameters remain challenging, particularly in the context of HSP. This paper introduces a comprehensive approach for gait analysis in patients with HSP, comprising two main components: stride segmentation utilizing a Hidden Markov Model (HMM), enabling robust identification of gait cycles from sensor data, and event detection and temporal gait parameter extraction. By leveraging the inherent temporal dynamics of gait patterns captured by wearable sensors, our method aims to overcome the limitations of existing techniques and provide reliable insights into the gait characteristics of HSP patients. Validation of this approach reveals promising results, with an F1 score of 89% for segmentation achieved through to-fold cross-validation. Additionally, our method demonstrates a mean absolute error of 0.008 seconds for stride time estimation compared to a gold standard motion capture system indicating the validity of our approach and its potential utility in hospital and real-world settings.

关键词： Accuracy Event detection Motion segmentation Hidden Markov models Motion capture Reliability Gait recognition Wearable sensors Parameter extraction Foot

Deep reinforcement learning for motion planning of mobile robots

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Butyrev, Leonid Edelhäußer, Thorsten Mutschler, Christopher Fraunhofer Institute for Integrated Circuits Iis Precise Positioning and Analytics Department Machine Learning and Information Fusion Group Nuremberg Germany Computer Science Department Machine Learning and Data Analytics Lab Erlangen Germany

This paper presents a novel motion and trajectory planning algorithm for nonholonomic mobile robots that uses recent advances in deep reinforcement learning. Starting from a random initial state, i.e., position, velocity and orientation, the robot reaches an arbitrary target state while taking both kinematic and dynamic constraints into account. Our deep reinforcement learning agent not only processes a continuous state space it also executes continuous actions, i.e., the acceleration of wheels and the adaptation of the steering angle. We evaluate our motion and trajectory planning on a mobile robot with a differential drive in a simulation environment. Copyright © 2019, The Authors. All rights reserved.

关键词： Deep learning

Hyperparameter Tuning MLP's for Probabilistic Time Series Forecasting

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Madhusudhanan, Kiran Jawed, Shayan Schmidt-Thieme, Lars Information Systems and Machine Learning Lab VWFS Data Analytics Research Center University of Hildesheim Hildesheim Germany

Time series forecasting attempts to predict future events by analyzing past trends and patterns. Although well researched, certain critical aspects pertaining to the use of deep learning in time series forecasting remain ambiguous. Our research primarily focuses on examining the impact of specific hyperparameters related to time series, such as context length and validation strategy, on the performance of the state-of-the-art MLP model in time series forecasting. We have conducted a comprehensive series of experiments involving 4800 configurations per dataset across 20 time series forecasting datasets, and our findings demonstrate the importance of tuning these parameters. Furthermore, in this work, we introduce the largest metadataset for time series forecasting to date, named TSBench, comprising 97200 evaluations, which is a twentyfold increase compared to previous works in the field. Finally, we demonstrate the utility of the created metadataset on multi-fidelity hyperparameter optimization tasks. © 2024, CC BY.

关键词： Forecasting

A Unified Approach Towards Active learning and Out-of-Distribution Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Schmidt, Sebastian Schenk, Leonard Schwinn, Leo Günnemann, Stephan BMW Group Munich Germany Technical University of Munich Data Analytics and Machine Learning Group Munich Germany

When applying deep learning models in open-world scenarios, active learning (AL) strategies are crucial for identifying label candidates from a nearly infinite amount of unlabeled data. In this context, robust out-of-distribution (OOD) detection mechanisms are essential for handling data outside the target distribution of the application. However, current works investigate both problems separately. In this work, we introduce SISOM as the first unified solution for both AL and OOD detection. By leveraging feature space distance metrics SISOM combines the strengths of the currently independent tasks to solve both effectively. We conduct extensive experiments showing the problems arising when migrating between both tasks. In these evaluations SISOM underlined its effectiveness by achieving first place in two of the widely used OpenOOD benchmarks and second place in the remaining one. In AL, SISOM outperforms others and delivers top-1 performance in three benchmarks. Copyright © 2024, The Authors. All rights reserved.

关键词： data handling

Robust Hyperbolic learning with Curvature-Aware Optimization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Bdeir, Ahmad Burchert, Johannes Schmidt-Thieme, Lars Landwehr, Niels Department of Data Analytics University of Hildesheim Germany The Information Systems and Machine Learning Lab University of Hildesheim Germany

Hyperbolic deep learning has become a growing research direction in computer vision due to the unique properties afforded by the alternate embedding space. The negative curvature and exponentially growing distance metric provide a natural framework for capturing hierarchical relationships between datapoints and allowing for finer separability between their embeddings. However, current hyperbolic learning approaches are still prone to overfitting, computationally expensive, and prone to instability, especially when attempting to learn the manifold curvature to adapt to tasks and different datasets. To address these issues, our paper presents a derivation for Riemannian AdamW that helps increase hyperbolic generalization ability. For improved stability, we introduce a novel fine-tunable hyperbolic scaling approach to constrain hyperbolic embeddings and reduce approximation errors. Using this along with our curvature-aware learning schema for Lorentzian Optimizers enables the combination of curvature and non-trivialized hyperbolic parameter learning. Our approach demonstrates consistent performance improvements across Computer Vision, EEG classification, and hierarchical metric learning tasks achieving state-of-the-art results in two domains and drastically reducing runtime. © 2024, CC BY.

关键词： Adversarial machine learning

Equivariant Representation learning for Augmentation-based Self-Supervised learning via Image Reconstruction

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Qin Krajsek, Kai Scharr, Hanno IAS-8: Data Analytics and Machine Learning Forschungszentrum Jülich Germany Forschungszentrum Jülich Germany

Augmentation-based self-supervised learning methods have shown remarkable success in self-supervised visual representation learning, excelling in learning invariant features but often neglecting equivariant ones. This limitation reduces the generalizability of foundation models, particularly for downstream tasks requiring equivariance. We propose integrating an image reconstruction task as an auxiliary component in augmentation-based self-supervised learning algorithms to facilitate equivariant feature learning without additional parameters. Our method implements a cross-attention mechanism to blend features learned from two augmented views, subsequently reconstructing one of them. This approach is adaptable to various datasets and augmented-pair based learning methods. We evaluate its effectiveness on learning equivariant features through multiple linear regression tasks and downstream applications on both artificial (3DIEBench) and natural (ImageNet) datasets. Results consistently demonstrate significant improvements over standard augmentation-based self-supervised learning methods and state-of-the-art approaches, particularly excelling in scenarios involving combined augmentations. Our method enhances the learning of both invariant and equivariant features, leading to more robust and generalizable visual representations for computer vision tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

Stream-based Active learning by Exploiting Temporal Properties in Perception with Temporal Predicted Loss

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Schmidt, Sebastian Günnemann, Stephan BMW Group Munich Germany Technical University of Munich Data Analytics and Machine Learning Group Munich Germany

Active learning (AL) reduces the amount of labeled data needed to train a machine learning model by intelligently choosing which instances to label. Classic pool-based AL requires all data to be present in a datacenter, which can be challenging with the increasing amounts of data needed in deep learning. However, AL on mobile devices and robots, like autonomous cars, can filter the data from perception sensor streams before reaching the datacenter. We exploited the temporal properties for such image streams in our work and proposed the novel temporal predicted loss (TPL)1 method. To evaluate the stream-based setting properly, we introduced the GTA V streets and the A2D2 streets dataset and made both publicly available. Our experiments showed that our approach significantly improves the diversity of the selection while being an uncertainty-based method. As pool-based approaches are more common in perception applications, we derived a concept for comparing pool-based and stream-based AL, where TPL outperformed state-of-the-art pool- or stream-based approaches for different models. TPL demonstrated a gain of 2.5 precept points (pp) less required data while being significantly faster than pool-based methods. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

FewSOME: One-Class Few Shot Anomaly Detection with Siamese Networks

学校读者我要写书评

暂无评论

FewSOME: One-Class Few Shot Anomaly Detection with Siamese N...

2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Belton, Niamh Hagos, Misgina Tsighe Lawlor, Aonghus Curran, Kathleen M. Science Foundation Ireland Centre for Research Training in Machine Learning Ireland University College Dublin School of Medicine Ireland University College Dublin School of Computer Science Ireland University College Dublin Insight Centre for Data Analytics Dublin Ireland

ISBN: (纸本)9798350302493

Recent Anomaly Detection techniques have progressed the field considerably but at the cost of increasingly complex training pipelines. Such techniques require large amounts of training data, resulting in computationally expensive algorithms that are unsuitable for settings where only a small amount of normal samples are available for training. We propose 'Few Shot anOMaly dEtection' (FewSOME), a deep One-Class Anomaly Detection algorithm with the ability to accurately detect anomalies having trained on 'few' examples of the normal class and no examples of the anomalous class. We describe FewSOME to be of low complexity given its low data requirement and short training time. FewSOME is aided by pretrained weights with an architecture based on Siamese Networks. By means of an ablation study, we demonstrate how our proposed loss, 'Stop Loss', improves the robustness of FewSOME. Our experiments demonstrate that FewSOME performs at stateof-the-art level on benchmark datasets MNIST, CIFAR-10, F-MNIST and MVTec AD while training on only 30 normal samples, a minute fraction of the data that existing methods are trained on. Moreover, our experiments show FewSOME to be robust to contaminated datasets. We also report F1 score and balanced accuracy in addition to AUC as a benchmark for future techniques to be compared against. Code available;https://***/niamhbelton/FewSOME. © 2023 IEEE.

关键词： Anomaly detection