Core: Spring 2019

You always start with the basics, and with Data Science it's no different! We'll be getting our feet wet with some simple, but powerful, models and demonstrate their power by applying them to real world data.

You've heard about them: Beating humans at all types of games, driving cars, and recommending your next Netflix series to watch, but what ARE neural networks? In this lecture, you'll actually learn step by step how neural networks function and how they learn. Then, you'll deploy one yourself!

Ever wonder how Facebook can tell you which friends to tag in your photos or how Google automatically makes collages and animations for you? This lecture is all about that: We'll teach you the basics of computer vision using convolutional neural networks so you can make your own algorithm to automatically analyze your visual data!

This lecture is all about Recurrent Neural Networks. These are networks with with added memory, which means they can learn from sequential data such as speech, text, videos, and more. Different types of RNNs and strategies for building them will also be covered. The project will be building a LSTM-RNN to generate new original scripts for the TV series “The Simpsons”. Come and find out if our networks can become better writers for the show!

Some of the hardest aspects of Machine Learning are the details. Almost every algorithm we use is sensitive to "hyperparameters" which affect the initialization, optimization speed, and even the possibility of becoming accurate. We'll cover the general heuristics you can use to figure out what hyperparameters to use, how to find the optimal ones, what you can do to make models more resilient, and the like. This workshop will be pretty "down-in-the-weeds" but will give you a better intuition about Machine Learning and its shortcomings.

In the fields of Data Science and Artificial Intelligence, your models and analyses will only be as good as the data behind them. Unfortunately, you will find that the majority of datasets you encounter will be filled with missing, malformed, or erroneous data. Thankfully, Python provides a number of handy libraries to help you clean and manipulate your data into a usable state. In today's lecture, we will leverage these Python libraries to turn a messy dataset into a gold mine of value!

Neural Nets are not the end all be all of Machine Learning. In this lecture, we will see how a decision tree works, and see how powerful a collection of them can be. From there, we will see how to utilize Random Forests to do digit recognition.

Support Vector Machines were among the most highly used ML algorithms before Neural Nets came back into the foreground. Unlike Neural Nets, SVMs can explain themselves quite well and allow us to use these ML mdels in fields like medicine, finance, and the like – where regulations require that we can inquire about our models.

We're filling this out!

We're filling this out!