Assessment remains a cornerstone of the educational process, with standardized testing often serving as a primary method for evaluating learning. However, as pedagogical approaches continue to evolve, Outcome-Based As...
详细信息
In this paper we initiate the study of expander decompositions of a graph G = (V, E) in the streaming model of computation. the goal is to find a partitioning C of vertices V such that the subgraphs of G induced by th...
详细信息
Kernel Extreme learning Machine (KELM) provides deterministic solution to pattern recognition problems in non-iterative manner. On contrary to ELM, the feature mapping is implicitly accomplished in KELM by practicing ...
详细信息
An integration of deep learning and symbolic logic is proposed, based on the Curry-Howard isomorphism and categorical logic. the propositional structure of logic is seen as a symmetry, namely the permutation invarianc...
详细信息
ISBN:
(纸本)9783030937577;9783030937584
An integration of deep learning and symbolic logic is proposed, based on the Curry-Howard isomorphism and categorical logic. the propositional structure of logic is seen as a symmetry, namely the permutation invariance of propositions;this can be implemented using so-called symmetric neural networks. Under our interpretation, it turns out that Google's BERT, which many currently state-of-the-art language models are derived from, can be regarded as an alternative form of logic. this BERT-like structure can be incorporated under a reinforcement-learning framework to form a minimal AGI architecture. We also mention some insights gleaned from category and topos theorythat point to future directions and may be helpful to other researchers, including mathematicians interested in AGI.
the goal of this tutorial is to provide the WSDM community with recent advances on the assessment and mitigation of data and algorithmic bias in recommender systems. We first introduce conceptual foundations, by prese...
详细信息
ISBN:
(纸本)9781450382977
the goal of this tutorial is to provide the WSDM community with recent advances on the assessment and mitigation of data and algorithmic bias in recommender systems. We first introduce conceptual foundations, by presenting the state of the art and describing real-world examples of how bias can impact on recommendation algorithms from several perspectives (e.g., ethical and system objectives). the tutorial continues with a systematic showcase of algorithmic countermeasures to uncover, assess, and reduce bias along the recommendation design process. A practical part then provides attendees with implementations of pre-, in-, and post-processing bias mitigation algorithms, leveraging open-source tools and public datasets;in this part, tutorial participants are engaged in the design of bias countermeasures and in articulating impacts on stakeholders. We conclude the tutorial by analyzing emerging open issues and future directions in this rapidly evolving research area (Website: https://***/wsdm2021).
Distributed denial of service attacks, in the recent times, have become a critical threat to not only private server connections but also meta scale cloud computing environments. this research paper furnishes a detail...
Distributed denial of service attacks, in the recent times, have become a critical threat to not only private server connections but also meta scale cloud computing environments. this research paper furnishes a detailed analysis of five machine learning algorithms in context with network intrusion detection that are; Isolation Forest, Random Forest, Support Vector Machine (SVM) and Artificial Neural Networks (ANN). the dataset used in this study, to differentiate between attack and normal cases, is the KDD cup 1999, wherein the dataset has been converted into binary classes to differentiate between the attack and normal cases. Following are the part of the research; Preprocessing the data, Scaling the features, Training the model and evaluating the performance using crucial metrics like accuracy, precision and F1 score and recall. Graphical representation and confusion matrices for data visualization have been made use of in addition to the aforementioned. the results show a penetrative information about the relative effectiveness and success of different ML algorithms in identifying network intrusions and hence can be used as insightful references for academics and practitioners in the field of network security. A well- explained algorithm selection, based on performance indicators enables an advanced approach to the study, thus providing a framework for further developments in systems of intrusion detection.
Despite some application of Contextual Teaching and learning (CTL) theories in practice, their effectiveness has not been prominent, necessitating comprehensive optimization. To enhance the quality and the success of ...
ISBN:
(数字)9781837241958
Despite some application of Contextual Teaching and learning (CTL) theories in practice, their effectiveness has not been prominent, necessitating comprehensive optimization. To enhance the quality and the success of typical processes like education activities, this study integrates CTL with Deming's PDCA (Plan-Do-Check-Act) cycle theory. through the stages of planning, implementation, and evaluation, the CTL model has been significantly improved. Methodologically, the study employs an inductive approach and experimental methods for data analysis. Results show significant improvements in the mentioned activities and enhances of performance and satisfaction of stakeholders like the population involved withthe optimized PDCA & CTL model. the results of this paper also confirmed the efficacy of the proposed method in process quality improvement .
In this article, a robust kernel extreme learning machine (KELM) framework is designed using mixture correntropy for recognition of facial images. KELM is augmentation of ELM with kernel learning concept, has attained...
详细信息
this systematic literature review examines predictive analytics and machine learning (ML) applications in edu-cational settings from 2019 to 2023. Machine learning (ML) algorithms are increasingly used to predict stud...
详细信息
ISBN:
(数字)9798331540883
ISBN:
(纸本)9798331540890
this systematic literature review examines predictive analytics and machine learning (ML) applications in edu-cational settings from 2019 to 2023. Machine learning (ML) algorithms are increasingly used to predict students' academic performance and are integral to early warning systems in educational settings. the study looks at several algorithms, from simple linear regression to more complex ensemble and neural network models. It also looks at how well they work in different types of schools and with different types of students. the analysis confirms that ML models provide educators with actionable insights to identify and support at-risk students. Along with evaluating predictive performance, the review discusses the ethical problems of using machine learning in education. these include worries about data privacy, the need for openness, and the possibility of bias in algorithmic decision-making. the paper concludes that while ML holds substantial promise in enhancing educational outcomes, its deployment requires prudent management to navigate the intersection of technological innovation and ethical practice, ensuring responsible use of data analytics in shaping the future of education.
the proceedings contain 36 papers. the special focus in this conference is on Artificial General Intelligence. the topics include: Goal Generation and Management in NARS;neuro-Symbolic Architecture for Experiential Le...
ISBN:
(纸本)9783030937577
the proceedings contain 36 papers. the special focus in this conference is on Artificial General Intelligence. the topics include: Goal Generation and Management in NARS;neuro-Symbolic Architecture for Experiential learning in Discrete and Functional Environments;Navigating Conceptual Space;A New Take on AGI;Categorical Artificial Intelligence: the Integration of Symbolic and Statistical AI for Verifiable, Ethical, and Trustworthy AI;moral Philosophy of Artificial General Intelligence: Agency and Responsibility;the Piagetian Modeler;epistolution: How a Systems View of Biology May Explain General Intelligence;measures of Intelligence, Perception and Intelligent Agents;Univalent Foundations of AGI are (not) All You Need;AGI Brain II: the Upgraded Version with Increased Versatility Index;biological Intelligence Considered in Terms of Physical Structures and Phenomena;adaptive Multi-strategy Market Making Agent;unsupervised Context-Driven Question Answering Based on Link Grammar;a Virtual Actor Behavior Model Based on Emotional Biologically Inspired Cognitive Architecture;causal Generalization in Autonomous learning Controllers;AI Future: From Internal Vectors to Simple Objects States Subspaces Maps;a thousand Brains and a Million theories;the Role of Bio-Inspired Modularity in General learning;the Ecosystem Path to AGI;on Comparative Analysis of Rule-Based Cognitive Architectures;elements of Task theory;20NAR1 - An Alternative NARS Implementation Design;the Gap Between Intelligence and Mind;neural String Diagrams: A Universal Modelling Language for Categorical Deep learning;AGI Control theory;AGI via Combining Logic with Deep learning;case-Based Task Generalization in Model-Based Reinforcement learning;pySigma: Towards Enhanced Grand Unification for the Sigma Cognitive Architecture;symbol Emergence and the Solutions to Any Task;compression, the Fermi Paradox and Artificial Super-Intelligence;the Artificial Scientist: Logicist, Emergentist, and Universalist Approaches t
暂无评论