Route selection for automated vehicle is a challenging task under critical situations, i.e., which can lead to serious or fatal injury. Due to the high risk of a dangerous collision with its consequences on human part...
Route selection for automated vehicle is a challenging task under critical situations, i.e., which can lead to serious or fatal injury. Due to the high risk of a dangerous collision with its consequences on human participants, the selection on possible routes poses ethical problems. This paper proposes a multi-layer route selection strategy for automated vehicles under critical situations, in which ethical considerations have been incorporated. The selection strategy contains a graph-based quantitative evaluation layer and a layer with qualitative evaluation based on the ethical principles. The operation of the route selection method through some examples is illustrated.
The prediction of terrain elevation values is a key task when it comes to off-road dynamics and inertial data estimation. A reliable elevation map can help in the estimation of future vehicle states and thus extend th...
The prediction of terrain elevation values is a key task when it comes to off-road dynamics and inertial data estimation. A reliable elevation map can help in the estimation of future vehicle states and thus extend the response time window for autonomous navigation and control. We trained a deep learning model that is able to successfully predict top-down terrain depth maps in an off-road setting using a lightweight monocular depth estimation network. The labels were generated using a custom preprocessing algorithm to aid single image depth model training. Unlike other elevation estimation algorithms, our work can predict terrain variation from a higher camera setting without the use of a multi-sensor system. The network is also shown to work outside of the training data domain. The code will be available at https://***/norbertmarko/terrain-depth.
In this study, we examined the use of computational techniques for accurately processing acoustic signals of human speech using digital media. Specifically, we focused on the Sanskrit language and applied a language m...
详细信息
In this study, we examined the use of computational techniques for accurately processing acoustic signals of human speech using digital media. Specifically, we focused on the Sanskrit language and applied a language modeling approach to improve recognition by machines. Our implementation of this approach for Sanskrit speech represents a novel approach in the field and has the potential to extract valuable information through automated processing. The ability to accurately process speech is crucial for effective human communication, and our research contributes to the development of more efficient and effective methods for achieving this goal..
In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between ...
In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between variables, and is also considered as a way to test for unfair discrimination in model predictions. Unexplored in previous literature, relaxing symmetry in Shapley values can have counter-intuitive consequences for model explanation. To better understand the method, we first show how local contributions correspond to global contributions of variance reduction. Using variance, we demonstrate multiple cases where ASV yields counter-intuitive attributions, arguably producing incorrect results for root-cause analysis. Second, we identify generalized additive models (GAM) as a restricted class for which ASV exhibits desirable properties. We support our arguments by proving multiple theoretical results about the method. Finally, we demonstrate the use of asymmetric attributions on multiple real-world datasets, comparing the results with and without restricted model families using gradient boosting and deep learning models.
The notion of lacunary infinite numerical sequence is introduced. It is shown that for an arbitrary linear difference operator L with coefficients belonging to the set R of infinite numerical sequences, a criterion (i...
Symmetric bi-manual manipulation is an essential skill in on-orbit operations due to its potent load capacity. Previous works have applied compliant control to maintain the stability of manipulations. However, traditi...
详细信息
The rapid growth of internet population poses a serious challenge to the security of internet resources. The security is directly affected by the hits of Denial of Services (DoS) attack which is rampant nowadays. With...
详细信息
The rapid growth of internet population poses a serious challenge to the security of internet resources. The security is directly affected by the hits of Denial of Services (DoS) attack which is rampant nowadays. With this evolving threat, designing a cutting-edge method is difficult from a cyber-security perspective. In this study, we propose a deep learning-based system for detecting Distributed Denial of Service (DDoS) attacks, which utilizes Logistic Regression, K- Nearest Neighbor, and Random Forest algorithms. We assess proposed models using a recently updated NSL KDD dataset. Our research’s findings also demonstrate that proposed model is highly accurate in detecting Distributed Denial of Service (DDoS) attacks. Our results show that our proposed model significantly improves upon current state-of-the-art attack detection methods
The article deals with the issues of building an emulator of a failed sensor, which is necessary to maintain the life cycle of a small spacecraft. Such an emulator is proposed to be built on the basis of a stable corr...
详细信息
In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between ...
详细信息
We introduce HunSum-1: a dataset for Hungarian abstractive summarization, consisting of 1.14M news articles. The dataset is built by collecting, cleaning and deduplicating data from 9 major Hungarian news sites throug...
详细信息
暂无评论