Machine learning, and deep learning in particular, has seen tremendous advances and surpassed human-level performance on a number of tasks. Currently, machine learning is increasingly integrated in many applications and thereby, becomes part of everyday life, and automates decisions based on predictions. In certain domains, such as medical diagnosis, security, autonomous driving, and financial trading, wrong predictions can have a significant influence on individuals and groups. While advances in prediction accuracy have been impressive, machine learning systems still can make rather unexpected mistakes on relatively easy examples, and the robustness of algorithms has become a reason for concern before deploying such systems in real-world applications. Recent research has shown that especially deep neural networks are susceptible to adversarial attacks that can trigger such wrong predictions. For image analysis tasks, these attacks are in the form of small perturbations that remain (almost) imperceptible to human vision. Such attacks can cause a neural network classifier to completely change its prediction about an image, with the model even reporting a high confidence about the wrong prediction. Of particular interest for an attacker are so-called backdoor attacks, where a specific key is embedded into a data sample, to trigger a pre-defined class prediction. In this paper, we systematically evaluate the effectiveness of poisoning (backdoor) attacks on a number of benchmark datasets from the domain of autonomous driving.
Exploration and exploitation of intelligent computing infrastructures are becoming of great interest for the research community to investigate different fields of science and engineering offering new improved versions of problem-solving soft computing-based methodologies. The current investigation presents a novel artificial neural network-based solution methodology for the presented problem addressing the properties of Hall current on magneto hydrodynamics (MHD) flow with Jeffery fluid towards a nonlinear stretchable sheet with thickness variation. Generalized heat flux characteristics employing Cattaneo–Christov heat flux model (CCHFM) along with modified Ohms law have been studied. The modelled PDEs are reduced into a dimensionless set of ODEs by introducing appropriate transformations. The temperature and velocity profiles of the fluid are examined numerically with the help of the Adam Bashforth method for different values of physical parameters to study the Hall current with Jeffrey fluid and CCHFM. The examination of the nonlinear input–output with neural network for numerical results is also conducted for the obtained dataset of the system by using Levenberg Marquardt backpropagated networks. The value of Skin friction coefficient, Reynold number, Deborah number, Nusselt number, local wall friction factors and local heat flux are calculated and interpreted for different parameters to have better insight into flow dynamics. The precision level is examined exhaustively by mean square error, error histograms, training states information, regression and fitting plots. Moreover, the performance of the designed solver is certified by mean square error-based learning curves, regression metrics and error histogram analysis. Several significant results for Deborah number, Hall parameters and magnetic field parameters have been presented in graphical and tabular form.
Case based learning technique (CBL) has been in practice throughout the world for a few decades now. Our institute adopted it some four years back, when it shifted towards modular system of teaching. It is the main technique being used for conducting small group discussions. We decided to introduce a new technique called the gamification technique of conducting small group discussion. Need was there to know the effectiveness of the new technique as well as to assess the factors for its preference so it could be modified to be more productive. The aim of this research was to assess the effectiveness of gamification teaching technique in comparison to the traditional CBL technique both quantitatively and qualitatively. This is a mixed‐method randomized controlled trial, conducted in Khyber Medical College on First Year Medical Students from June to October 2021. Group based teaching involving both CBL and gamification approach was used in this study in a cross over manner. Addressing the ethical concerns, and after informed consent pre‐ and post‐testing was done to quantify the performance and an open‐ended survey was disseminated after the sessions to check the perception of the students. The study recorded (quantitatively) that post testing mean score of gamification teaching technique was 3.41±0.982, while for CBL the mean was 3.55±1.055, recording an insignificant difference with p‐value of 0.608. In qualitative analysis, about 12 (80%) students preferred the gamification technique, their perception being that it instills competitiveness, increases the involvement of students in class and their motivation level. This research further revealed that the CBL approach had the advantage of quick learning via the facilitator presentation, and due to the handouts, it was easy to follow. Some of the negative points of CBL reported were that; the participants found it a boring and monotonous way of learning. The drawback of gamification reported was, that the students were unsure about the accuracy of the information they prepared initially as it wasn’t being directed by the facilitator. The study concluded an insignificant difference among the two techniques quantitatively; however, qualitatively the students preferred gamification as a better technique for learning.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.