Classification
link to mnist
This page is for finding a classifier on the KMNIST dataset. This dataset is more challenging than the original MNIST dataset that I have previously solved.
The details of the dataset can be found in the associated paper.
In short, since the reformation of the Japanese education in 1868, there became a standardisation of the kanji characters, and in the present day, most Japanese people cannot read the texts from 150 years ago.
Here we aim to understand how exactly "Logistic Regression", a method that seems only to be used for classification problems, is indeed a regression algorithm.
We will as per the trend thus far, detail closed-form and approximate solutions to the loss function on this page. Furthermore, we will see how this type of regression is still a member of the GLM (generalised linear models) family, and we shall witness the derivation of the loss function by assuming our data is Bernoulli distributed.
History
Abstract
The MNIST dataset (Modified National Institute of Standards and Technology) has been very influential in machine learning and computer vision. It is an easy and popular dataset that has been used since it's inception in 1998 as a benchmark for Machine Learning Models. Historically it has enhanced the evolution of OCR (Optical Character Recognition) and assisted in the emergence of neural networks.
Origins
The story of MNIST begins with the NIST dataset, developed by the United States National Institute of Standards and Technology in the late 1980s. The original dataset was created to facilitate research in OCR systems, which were becoming increasingly relevant for automating tasks like check processing and mail sorting. NIST's dataset consisted of tens of thousands of handwritten digits collected from various sources, including Census Bureau employees and high school students.
Origins
The perceptron learning algorithm is the most simple algorithm we have for Binary Classification.
It was introduced by Frank Rosenblatt in his seminal paper: "The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain" in 1958. The history however dates back further to the theoretical foundations of Warren McCulloch and Walter Pitts in 1943 and their paper "A Logical Calculus of the Ideas Immanent in Nervous Activity". The interested reader may visit these links for annotations and the original pdfs.
Might not actually be Kaggle, please check.