kconner

macOS Internals

Understand your Mac and iPhone more deeply by tracing the evolution of Mac OS X from prelease to Swift. John Siracusa delivers the details.

Starting Points

How to use this gist

You've got two main options:

yoavg

Reinforcement Learning for Language Models

Yoav Goldberg, April 2023.

Why RL?

With the release of the ChatGPT model and followup large language models (LLMs), there was a lot of discussion of the importance of "RLHF training", that is, "reinforcement learning from human feedback". I was puzzled for a while as to why RL (Reinforcement Learning) is better than learning from demonstrations (a.k.a supervised learning) for training language models. Shouldn't learning from demonstrations (or, in language model terminology "instruction fine tuning", learning to immitate human written answers) be sufficient? I came up with a theoretical argument that was somewhat convincing. But I came to realize there is an additional argumment which not only supports the case of RL training, but also requires it, in particular for models like ChatGPT. This additional argument is spelled out in (the first half of) a talk by John Schulman from OpenAI. This post pretty much

yoavg

Some remarks on Large Language Models

Yoav Goldberg, January 2023

Audience: I assume you heard of chatGPT, maybe played with it a little, and was imressed by it (or tried very hard not to be). And that you also heard that it is "a large language model". And maybe that it "solved natural language understanding". Here is a short personal perspective of my thoughts of this (and similar) models, and where we stand with respect to language understanding.

Intro

Around 2014-2017, right within the rise of neural-network based methods for NLP, I was giving a semi-academic-semi-popsci lecture, revolving around the story that achieving perfect language modeling is equivalent to being as intelligent as a human. Somewhere around the same time I was also asked in an academic panel "what would you do if you were given infinite compute and no need to worry about labour costs" to which I cockily responded "I would train a really huge language model, just to show that it doesn't solve everything!". We

Spuffynism

Solving a Dungeons & Dragons riddle using prolog

Bringing back the magic of Christmas using the magic of prolog

As part of a holiday D&D one-shot session where Santa Claus's toy factory had been sabotaged, our dungeon master presented to us, a group of Christmas elves, a riddle to solve.

9 cards, labeled with the names of Santa's reindeer were presented to us. The instructions indicated that we had to find the order reindeer were in, according to this riddle:

Vixen should be behind Rudolph, Prancer and Dasher, whilst Vixen should be in front of Dancer and Comet. Dancer should be behind Donder, Blitzen and Rudolph. Comet should be behind Cupid, Prancer and Rudolph. Donder should be behind Comet, Vixen, Dasher, Prancer and Cupid. Cupid should be in front of Comet, Blitzen, Vixen, Dancer and Rudolph. Prancer should be in front of Blitzen, Donder and Cupid. Blitzen should be behind Cupid but in front of Dancer, Vixen and Donder. Rudolph should be behind Prancer but in front of Dasher, Dancer and Dond

hamelsmu

nofreewill42

radekosmulski

Summary of Fastai.ai's in-depth discussion of types of normalization

# simulated batch of images
x = torch.rand(64, 3, 224, 224)
# or some number of layers up the convolutional stack
x = torch.rand(64, 256, 32, 32)

BatchNorm:

rxwei

See the official Differentiable Programming Manifesto instead.

abridgland

shagunsodhani

Conditional Generative Adversarial Nets

Introduction

• Conditional version of Generative Adversarial Nets (GAN) where both generator and discriminator are conditioned on some data y (class label or data from some other modality).
• Link to the paper

Architecture

• Feed y into both the generator and discriminator as additional input layers such that y and input are combined in a joint hidden representation.