-
A privacy-preserving protocol for age-verified web applications
Poorly-designed bills have recently passed in California, Colorado, New York, and Illinois, requiring intrusive, privacy-compromising age verification procedures for sites online. The motivation is to prevent children and adolescents accessing pornography and social media. However, the ill-considered means of achieving this may be worse than doing nothing at all: by forcing users to disclose their…
-
On the products of Bernoulli random variables
I took another detour into the world of binary neural networks. In the interests of evaluating dot-product-like functions of two binary vectors, I wanted to assess how the distributions of multiplied bits (equivalent to ANDed bits) relate to the distributions of the individual bits. This post captures the results of this side-trip. Products of independent…
-
A normally-distributed position encoding
As part of my effort to develop a probabilistic interpretation of transformer language models, I became interested in alternative position encodings to that used in Attention Is All You Need. A position encoding can be characterized as some function from a non-negative integer to a vector of reals: $$e : \mathbb{N} \to \mathbb{R}^{K}$$ or as…
-
Concept: Rational Reader
This is a sketch of a solution to Task: text to Bayes rationality. The paradigm is Bayesian epistemology. The broader task is to infer a rational worldview from empirical observation. Here, we use a collection of documents as our link to the real world: we observe that somebody created a document like this. Roughly speaking,…
-
Task: text to Bayes rationality
Language models, at the core, are very stupid, blindly predicting the next word given the preceding words. This leaves them profoundly vulnerable to the biases and inaccuracies of the training data. Human annotations are applied late in the game to reduce spurious, hallucinatory, extremist, discriminatory, and other undesired outputs, but this after-the-model reshaping is symptomatic…
-
Task: 3d reconstruction
https://paperswithcode.com/task/3d-reconstruction
-
Dataset: Wikimedia Commons Image Super-Resolution
TODO full description TODO make available on torrent? A large dataset of lossless images from Wikimedia Commons, with 25%, 50%, and 128×128 downscales, plus train and validation splits for an image super-resolution task.
-
Task: image super-resolution
https://paperswithcode.com/task/image-super-resolution TODO link my approach TODO link my dataset
-
Task: image-text joint embedding
A single vector that has a sensible text and image representation. TODO link my approach in https://github.com/joshhansen/MLPortfolio Related:
-
Task: image to video
Given a frame, produce video as would occur in a movie https://paperswithcode.com/task/image-to-video