Artificial Intelligence

2408 Submissions

[6] viXra:2408.0130 [pdf] replaced on 2024-09-11 13:41:25

Bayesian Networks, Kullback-Leibler and Topology

Authors: Ait-Taleb nabil
Comments: 5 Pages.

In this paper, I will propose a topology allowing to measure a neighborhood for the Bayesian networks.This topology will correspond to a Kullback-Leibler distance ratio and will allow to know the distance between a current Bayesian network and a Bayesian network having a chain rule. This topology applied to Bayesian networks will be normalized and will therefore vary from 0 to 1. The value 0 will correspond to a Bayesian network with a chain rule and the value 1 to a Bayesian network without edges.
Category: Artificial Intelligence

[5] viXra:2408.0124 [pdf] submitted on 2024-08-28 20:50:31

Abstractive Text Summarisation Using T5 Transformer Architecture with Analysis

Authors: Vasanth Kumar Bhukya, Umesh Bhukya
Comments: 22 Pages. 20 figures, 6 chapters

Now a days, Text summarization has become important as the amount of text data available online grows at an exponential rate. Most of the text classification systems require going through a huge amount of data. In general,Producing exact and meaningful summaries of big texts is a time-consuming endeavour. Hence generating abstract summaries which retain the key information of the data and using it to train machine learning models will makethese models space and time-efficient. Abstractive text summarization has beensuccessful in moving from linear models to nonlinear neural network models using sparse models [1]. This success comes from the application of deep learning models on natural language processing tasks where these mod-els are capable of modeling the interrelating patterns in data without hand-crafted features. The Text to Text Transfer Transformer(T5) approach was used to investigate the text summarization problem, and the results showed that the Transfer Learning based model performed significantly better for abstractive text summarization than the Sequence to Sequence Recurrent Model.
Category: Artificial Intelligence

[4] viXra:2408.0118 [pdf] submitted on 2024-08-27 05:40:26

Graph Neural Network for Molecular Structure: Application in HIV Inhibitor Molecule Prediction

Authors: Quynh Nguyen
Comments: 14 Pages.

The application of Graph Neural Networks (GNNs) in computational chemistry provides a powerful approach to modeling and predicting the properties of molecular compounds. GNNs represent atoms as nodes and bonds as edges, capturing the complex interactions within molecular graphs. This approach offers a robust method for predicting chemical properties, including molecular stability, reactivity, and toxicity. In this paper, we explore various GNN architectures and their ability to generalize across different molecular datasets, such as QM9 and MoleculeNet. As a specific application, we propose a novel framework that utilizes GNNs to predict and identify potential HIV inhibitor molecules by analyzing their graph-based representations. This research aims to contribute to the discovery and design of effective HIV inhibitors, offering a promising direction for future antiviral drug development.
Category: Artificial Intelligence

[3] viXra:2408.0087 [pdf] replaced on 2025-08-22 16:53:54

How Can We Make AI with a Nice Character?

Authors: Dimiter Dobrev, Lyubomir Ivanov, George Popov, Vladimir Tzanov
Comments: 24 Pages.

God created man in His own image, the Bible said millennia ago. Today we are headed to creating Artificial Intelligence (AI) in our own image. The difference however is that God created a feeble and vulnerable being for which to take care of, while we are trying to create an almighty being who will be incomparably smarter than us and will take care of us. Thus, we are aiming to create our new god, and it matters a lot what kind of character the new god will be — kind and compassionate, or terribly stringent and overly demanding on us. Every human being has a character. Similarly, AI will have its own character. We will consider AI as a program with parameters which determine its character. The aim is to use these parameters in order to define the kind of character we want AI to have.
Category: Artificial Intelligence

[2] viXra:2408.0038 [pdf] submitted on 2024-08-09 16:14:18

Difference Between the Notion of Causation and Pearson Correlation in a Multivariate Gaussian Context

Authors: Ait-Taleb nabil
Comments: 11 Pages.

In a Gaussian multivariate context, we will describe the steps to follow to differentiate the notion of Pearson correlation and the causality. This paper includes numerical examples clearly showing the difference between the two notions.
Category: Artificial Intelligence

[1] viXra:2408.0037 [pdf] submitted on 2024-08-09 19:36:22

Breast Cancer Segmentation in Medical Imaging: A Custom U-Net Approach

Authors: Tanvir Rahman, Ataur Rahman, Tamanna Afroz, Rafia Akhter
Comments: 6 Pages.

Deep learning, particularly using U-Net architecture, has shown remarkable performance in various image segmentation tasks, including medical and non-medical applications. This versatile approach enables automated analysis of complex images, which is crucial for improving diagnostic accuracy and efficiency. For medical applications, breast cancer detection serves as a prominent example, where deep learning models have demonstrated superior performance over traditional methods. We examine various techniques used to enhance U-Net's ability to detect breast cancer, Moreover, we review the most commonly used datasets for medical image segmentation tasks effectiveness in a range of applications. Our proposed custom U-Net model extends the standard U-Net architecture by incorporating advanced techniques to enhance its ability to handle segmentation tasks. These improvements result in improved accuracy, Intersection over Union (IOU) scores, and dice coefficient scores, setting a new benchmark forsegmentation models.
Category: Artificial Intelligence