/explore

Click through on any links that interest you or select the planets on the right to continue exploring the Outer Web.
You are here

 www.lesswrong.com
RL with KL penalties is better seen as Bayesian inference - LessWrong
| | www.danieldjohnson.com
Uncertain Simulators Don't Always Simulate Uncertain Agents | Daniel D. Johnson
3.4 parsecs away

Travel
| | I argue that hallucinations are a natural consequence of the language modeling objective, which focuses on simulating confident behavior even when that behavior is hard to predict, rather than predictable behaviors that take uncertainty into account. I also discuss five strategies for avoiding this mismatch.
| | www.depthfirstlearning.com
Variational Inference with Normalizing Flows ยท Depth First Learning
2.5 parsecs away

Travel
| | [AI summary] The user has provided a detailed and complex set of questions and reading materials related to normalizing flows, variational inference, and generative models. The content covers topics such as the use of normalizing flows to enhance variational posteriors, the inference gap, and the implementation of models like NICE and RealNVP. The user is likely seeking guidance on how to approach these questions, possibly for academic or research purposes.
| | jxmo.io
Introduction to variational autoencoders - Jack Morris
3.4 parsecs away

Travel
| | A primer on variational autoencoders (VAEs) culminating in a PyTorch implementation of a VAE with discrete latents.
| | wtfleming.github.io
Cats vs Dogs - Part 2 - 98.6% Accuracy - Binary Image Classification with Keras and Transfer Learning
9.6 parsecs away

Travel
| [AI summary] This article demonstrates how to achieve 98.6% accuracy in binary image classification of cats and dogs using Keras and transfer learning with the VGG16 model.