Kolmogorov Arnold Networks (KANs)
Exploring the Theoretical Foundations and Applications
Posted on August 26, 2024
A new research called “KAN: Kolmogorov–Arnold Network” has created excitement in the machine learning (ML) community. It presents a new perspective on Neural Networks, being a possible alternative to Multi-Layer Perceptrons (MLP), a cornerstone of current ML [1]. Information Original...
Stop Oversampling for Class Imbalance Learning
Why oversampling methods should be avoided, especially in sensitive applications
Posted on March 29, 2024
Information Paper link: https://arxiv.org/pdf/2202.03579.pdf Published date: June 8th, 2022 Authors: Ahmad B. Hassanat , Ahmad S. Tarawneh , Ghada A. Altarawneh , Abdullah Almuhaimeed Purpose Analysis of a large number of oversampling techniques and devised a new oversampling evaluation system...
Top 10 hardest mathematical problems that remain unsolved
Math problems that have been challenging us for centuries
Posted on February 28, 2024
Despite recent breakthroughs like a supercomputer cracking the Sum of Three Cubes problem after 65 years, mathematicians continue to grapple with centuries-old challenges, reminding us that even the toughest math problems may eventually yield to human perseverance, or perhaps not....
The Rise of the AI Engineer
One can be quite successful in this role without ever training anything!
Posted on February 27, 2024
There are an increasing in the Models, tools and different applications that AI can have. “In numbers, there’s probably going to be significantly more AI Engineers than there are ML engineers / LLM engineers. One can be quite successful in...
Molecular digital data storage using DNA
Exploring DNA as a Cutting-Edge Solution for Digital Data Storage
Posted on February 27, 2024
Molecular data storage presents a viable solution to the increasing need for dense and durable information storage. DNA exemplifies effective archival data storage in molecular form. The review outlines the process, current advancements, and challenges for\ mainstream adoption. Additionally, it...
Extreme Learning Machine (ELM)
Using Moore Penrose generalized inverse to set weights
Posted on February 27, 2024
It is a feedforward neural network that uses Moore Penrose generalized inverse to set its weights. ELMs are able to produce acceptable predictive performance and it learn thousands of times more faster than other algorithm and their computational cost is...
Dunning Kruger Effect
Cognitive bias in which people believe they are smarter and more capable than they are
Posted on February 27, 2024
The Dunning-Kruger effect is a type of cognitive bias in which people believe they are smarter and more capable than they are. Essentially, low-ability people do not possess the skills needed to recognize their own incompetence. The combination of poor self-awareness and...
Pydantic
Simplifying Python Data Validation, manipulation and Serialization
Posted on February 26, 2024
Definition Pydantic is a Python package that can offer simple data validation and manipulation. It was developed to improve the data validation process for developers. Indeed, Pydantic is an API for defining and validating data that is flexible and easy...
FastAI
Simple Solutions for Complex Challenges
Posted on February 26, 2024
FastAI is a user-friendly deep learning library that simplifies complex machine learning tasks. It provides pre-built models and tools to make it easier for beginners to create powerful AI applications without needing extensive coding knowledge. Higlights: In general it is...
Bongard Problems
Distinguishing patterns between two sets of simple shapes.
Posted on February 26, 2024
Definition Bongard problems are a series of logic puzzles designed by the Russian researcher and computer scientist, K. M. Bongard, in the late 1960s. The puzzles consist of two panels of simple, abstract computer-generated images. A participant is asked to...