Published in Towards Data Science·Oct 5, 2018An Intuitive Explanation of Policy GradientThis is part 1 of a series of tutorials which I expect to have 2 or 3 parts. The next part will be on A2C and, time providing, I hope to complete a part on various forms of off-policy policy gradients. The notebooks for these posts can be found in…Machine Learning13 min readMachine Learning13 min read
Published in Towards Data Science·May 5, 2018Deep Learning Book Notes, Chapter 3 (part 1): Introduction to ProbabilityThese are the first part of my notes for chapter 3 of the Deep Learning book. They can also serve as a quick intro to probability. These notes cover about half of the chapter (the part on introductory probability), a followup post will cover the rest (some more advanced probability…Machine Learning18 min readMachine Learning18 min read
Published in Towards Data Science·Apr 2, 2018Paper repro: Deep Metalearning using “MAML” and “Reptile”In this post I reproduce two recent papers in the field of metalearning: MAML and the similar Reptile. The full notebook for this reproduction can be found here. The goal of both of these papers is to solve the K-shot learning problem. In K-shot learning, we need to train a…Machine Learning8 min readMachine Learning8 min read
Published in Towards Data Science·Mar 27, 2018Paper Repro: Deep NeuroevolutionIn this post, we reproduce the recent Uber paper “Deep Neuroevolution: Genetic Algorithms are a Competitive Alternative for Training Deep Neural Networks for Reinforcement Learning”, which amazingly showed that simple genetic algorithms sometimes performed better than apparently advanced reinforcement learning algorithms on well studied problems such as Atari games. We…Machine Learning20 min readMachine Learning20 min read
Published in Becoming Human: Artificial Intelligence Magazine·Mar 17, 2018Deep Learning Book Notes, Chapter 2: Linear Algebra for Deep LearningThese are my notes for chapter 2 of the Deep Learning book. They can also serve as a quick intro to linear algebra for deep learning. My notes for chapter 1 can be found below: Deep Learning Book Notes, Chapter 1 These are my notes on the Deep Learning book. There are many like them but these ones are mine.becominghuman.ai For this section I decided to make things a bit more intuitive using code, which…Machine Learning17 min readMachine Learning17 min read
Published in Becoming Human: Artificial Intelligence Magazine·Mar 11, 2018Paper repro: “Self-Normalizing Neural Networks”This post details my experience reproducing “Self-Normalizing Neural Networks” as part of the Nurture.AI NIPS Challenge 2017, as well as my experience participating in the challenge. My notebook can be found here. The goal of the NIPS challenge 2017 was to provide code implementation for various NIPS papers as well…Machine Learning10 min readMachine Learning10 min read
Published in Becoming Human: Artificial Intelligence Magazine·Mar 10, 2018Deep Learning Book Notes, Chapter 1These are my notes on the Deep Learning book. There are many like them but these ones are mine. They are all based on my second reading of the various chapters, and the hope is that they will help me solidify and review the material easily. …Artificial Intelligence8 min readArtificial Intelligence8 min read
Published in Becoming Human: Artificial Intelligence Magazine·Mar 6, 2018Investigating Focal and Dice Loss for the Kaggle 2018 Data Science BowlThis post details my experiments and implementations with three important loss functions for the Kaggle 2018 data science bowl, and compares their effects on a simplified implementation of U-Net. To those unfamiliar, the 2018 Data Science Bowl is a nuclei extraction competition. It consists in finding which pixels in images…Machine Learning7 min readMachine Learning7 min read
Published in Becoming Human: Artificial Intelligence Magazine·Feb 27, 2018Paper repro: “Learning to Learn by Gradient Descent by Gradient Descent”This post is the first of what will (hopefully) be a series of deep learning paper reproduction posts. Important note: the primary goal of this post is not to best or first tutorial on a given paper, but simply to keep a record of the papers I have reproduced and…Deep Learning10 min readDeep Learning10 min read
Published in Becoming Human: Artificial Intelligence Magazine·Oct 31, 2017Beat Atari with Deep Reinforcement Learning! (Part 2: DQN improvements)Note: Before reading part 2, I recommend you read Beat Atari with Deep Reinforcement Learning! (Part 1: DQN) Finally, part 2 is here! Training DQNs can take a while, especially as you get closer to the state of the art. …Machine Learning8 min readMachine Learning8 min read