Since its first documented appearance about four millennia ago, the number π (pi) has become not just an oft-encountered constant in mathematics and physics but a cultural phenomenon, with numerous appearances in books, artworks, TV shows, and films. This talk will discuss a handful of these somewhat recreational aspects of pi, each of which turns out to have non-trivial connections to serious mathematics and/or computer algorithms. Among our topics will be curious features of pi's digits, memorizing the digits using a human-based algorithm, pi in graphic design, the most beautiful digit-computing algorithm, and the wonders of Pilish: a constrained dialect of English in which the number of letters in successive words matches the digits of pi. Speaker(s): Michael Keith Room: 105, Bldg: Computer Science Building, 35 Olden St, Princeton, New Jersey, United States, 08544, Virtual: https://events.vtools.ieee.org/m/439050

Course Format: Live Webinar, 4.0 hours of instruction! Series Overview: From the book introduction: “Neural networks and deep learning currently provides the best solutions to many problems in image recognition, speech recognition, and natural language processing.” This Part 1 and the planned Part 2 (to be confirmed) series of courses will teach many of the core concepts behind neural networks and deep learning. This is a live instructor-led introductory course on Neural Networks and Deep Learning. It is planned to be a two-part series of courses. The first course is complete by itself and covers a feedforward neural network (but not convolutional neural network in Part 1). It will be a pre-requisite for the planned Part 2 second course. The class material is mostly from the highly-regarded and free online book “Neural Networks and Deep Learning” by Michael Nielsen, plus additional material such as some proofs of fundamental equations not provided in the book. More from the book introduction: Reference book: “Neural Networks and Deep Learning” by Michael Nielsen, http://neuralnetworksanddeeplearning.com/ “We’ll learn the core principles behind neural networks and deep learning by attacking a concrete problem: the problem of teaching a computer to recognize handwritten digits. …it can be solved pretty well using a simple neural network, with just a few tens of lines of code, and no special libraries.” “But you don’t need to be a professional programmer.” The code provided is in Python, which even if you don’t program in Python, should be easy to understand with just a little effort. Benefits of attending the series: * Learn the core principles behind neural networks and deep learning. * See a simple Python program that solves a concrete problem: teaching a computer to recognize a handwritten digit. * Improve the result through incorporating more and more core ideas about neural networks and deep learning. * Understand the theory, with worked-out proofs of fundamental The demo Python program (updated from version provided in the book) can be downloaded from the speaker’s GitHub account. The demo program is run in a Docker container that runs on your Mac, Windows, or Linux personal computer; we plan to provide instructions on doing that in advance of the class. (That would be one good reason to register early if you plan to attend, in order that you can receive the straightforward instructions and leave yourself with plenty of time to prepare the Git and Docker software that are widely used among software professionals.) Course Background and Content: This is a live instructor-led introductory course on Neural Networks and Deep Learning. It is planned to be a two-part series of courses. The first course is complete by itself and covers a feedforward neural network (but not convolutional neural network in Part 1). It will be a pre-requisite for the planned Part 2 second course. The class material is mostly from the highly-regarded and free online book “Neural Networks and Deep Learning” by Michael Nielsen, plus additional material such as some proofs of fundamental equations not provided in the book. Outline: - Feedforward Neural Networks - Simple (Python) Network to classify a handwritten digit - Learning with Stochastic Gradient Descent - How the backpropagation algorithm work - Improving the way neural networks learn: - - Cross-entropy cost function - SoftMax activation function and log-likelihood cost function - Rectified Linear Unit - Overfitting and Regularization: - - L2 regularization - Dropout - Artificially expanding data set Pre-requisites: There is some heavier mathematics in learning the four fundamental equations behind backpropagation, so a basic familiarity with multivariable calculus and matrix algebra is expected, but nothing advanced is required. (The backpropagation equations can be also just accepted without bothering with the proofs since the provided Python code for the simple network just make use of the equations.) Basic familiarity with Python or similar computer language. Speaker(s): CL Kim, Boston, Massachusetts, United States, Virtual: https://events.vtools.ieee.org/m/450631