Viewing profile — dustintran

Hello. I'm the person that was linked to in that GitHub issue! I sympathize with the post's frustration. The TF tutorials on the official website are well-written. But they mostly …

The author doesn't really apply Bayesian inference. He applies Bayes' rule which is a mathematical property.

Note that this should not validate or invalidate his claims, but here are videos of his teaching: http://math1afall2015.blogspot.com

You've voiced my own feelings exactly.

The paper described in the article: http://arxiv.org/pdf/1506.05439v1.pdf

Hi, stan dev here. I think viewing Stan as a better BUGS is helpful but limiting. The syntax is similar, but the class of models Stan fits is far more general. The class of algorit…

Most all probabilistic programming languages in fact treat every model as equivalent to an HMM. So certainly inference on them can be done.

The conundrum during all research talks.

To clarify, it has been studied in Zoubin Ghahramani's group [1] (and also more recently in Ryan Adam's group [2]), and it's most widely known through Radford Neal [3] who's won a …

This is not completely accurate. Deep learning at the moment simply means the use of "deep" architectures in neural networks. Graphical models, standard Bayesian hierarchical model…

I strongly disagree with not using linear models, at least to build some theory and intuition before continuing with more sophisticated algorithms. What I find to be more egregious…

I'm curious to see when spam bots will eventually use sophisticated language models in order to generate text, aimed to combat a search engine's own machine learning algorithms whi…

The problem is the more than likely chance that only doing one resample (in bootstrap terminology) may not lead to any clear rejection of statistical significance. That is, it is q…

This is not a good analogy. Money is the obvious reason one stays at the "lower hill". Thus it is not a question of whether the person can stoop down from the "lower hill" and get …

Yup, it seems k was fixed since the first time these scripts were made for NIPS 2012 (?). Some of the more well-established advances since LDA would also likely help, like HDP.

I imagine more preventative solutions to develop earlier which are more effective, e.g., "treat" citizens who are predicted highly likely to be cause a dangerous outcome in the fut…

You can split screens (and virtually never use the mouse) in Vim. What particular feature are you looking for here?

I'm not sure how useful this would be. Do people willfully drive from work to a coffee shop in order to get work done, and not simply walk to their favorite coffee shop within walk…

Shameless hijacking and also just not as good..