A heuristic for computing similarity in unstructured image data. — Introduction How could we compute how similar one image is to another? For similarity among data in a vectorized form, we can find the sum of the squared differences between two examples, or use similar methods like cosine similarity. However, performing such techniques on images — summing the squared difference between…

Here is the organized index of my series of philosophical articles, Paradox & Refutations, and other miscellanea on the central questions of life and logic. Each shapes the strange philosophy of what makes us us. On the Central Questions Ultimate Meaning: Its nonexistence, and good riddance

Introduction Last post reasoned using the the almost-whimsical teletransportation thought experiment. Here, we will reify that rarefied philosophy with the real — yet oh so strange — corpus callosotomy surgery, and in doing so, plumb deeper into the philosophy of mind and metaphysics of identity. Problem Patient P suffers from epilepsy, intense…

Introduction The Mars Teletransporter, introduced by Derek Parfit[9], is a paradoxical and philosophically trying beast whose resolution requires that we dive into and criticize current philosophical frameworks. As the argument reaches its terminus, we will have developed and tested a counterintuitive explanation of who we are that can bear up to…

Abstract: Does a ship remain the same ship if all its planks are replaced with new planks gradually? Yes. How about if the renovation was sudden? No. The stance may be puzzling, but it is consistent. The locus of identity is in our brains, not the external world. We define…

Abstract Some philosophers have been shifting the discussion from meaning of life to meaning in life. Not so fast. There are things to learn about the former — it doesn’t exist — and how it impacts us (positively). Let’s define meaning as a point, an end that justifies an endeavor. A…

Knowledge: Limits and Potential I’ve heard of parents who try to take their child’s questions about the world seriously. The most sincere ones become distraught. …

I wanted an easy way to get a state-of-the-art image augmentation pipeline with no manual iteration, no separate models to train and no thinking. To provide that, I created MuarAugment (Model Uncertainty- And Randomness-based Augmentation), a GPU-supported Python package built on Pytorch, Albumentations and Kornia. This article guides you through…

Give them a try: they perform better than one-hot encodings. — This post is a guide for using embeddings to represent categorical variables in machine learning algorithms. I’ve seen other articles address this method, but none actually showing how to do it in code. So I decided to fill that lacuna with my own post. If you are itchy for the…