kianN

8 ms·

61.
by kianN 1y ago
The URL is unique to your search and saves it's state! In the technical notes I sort of laid out our model graph on the document branch. We also have a topic branch that is also structured hierarchically: Uniform Prior → High Level Top
62.
by kianN 1y ago
Some statistical notes for those interested: Under the hood, this model resembles LDA, but replaces its Dirichlet priors with Pitman–Yor Processes (PYPs), which better capture the power-law behavior of word distributions. It also supports
63.
Show HN: Research Hacker News, ArXiv & Google with Hierarchical Bayesian Models(sturdystatistics.com)
85 points by kianN 1y ago|23 comments
64.
by kianN 1y ago
Additional link from Reuters: https://www.reuters.com/business/microsoft-openai-reach-new-...
65.
by kianN 1y ago
It’s a pretty complex topic with a lot of variance. I think an accumulation of detailed examples is the first step towards empirical data.
66.
Spotting AI Articles(maurycyz.com)
4 points by kianN 1y ago|2 comments
67.
by kianN 1y ago
Haha that’s embarrassing! The progress bars are an estimate. If a paper has a lot of citations, it may take a bit longer than the duration of the bars but it will hopefully finish relatively soon! Edit: Got home and checked the error logs.
68.
by kianN 1y ago
If you’re interested in a literature review tool, I built a public one for some friends in grad school that uses hierarchical mixture models to organize bulk searches and citation networks. Example: https://platform.sturdystatist
69.
Do NFL Referees Favor the Kansas City Chiefs?(brian-lookabaugh.github.io)
3 points by kianN 1y ago|0 comments
70.
Hypernetworks: Neural Networks for Hierarchical Data(blog.sturdystatistics.com)
2 points by kianN 1y ago|0 comments
71.
by kianN 1y ago
Hi author here. The goal of this post was to do a data driven analysis on the investment trends of companies that are leveraging LLMs. I used YC company descriptions to get a granular sense of the use cases in which LLMs are gaining and los
72.
AI Keeps Shifting Right: Coping with the Limitations of Large Language Models(blog.sturdystatistics.com)
3 points by kianN 1y ago|1 comments
73.
by kianN 1y ago
Haha I’ve actually never heard of that field. My work was focused on applying Chinese restaurant process models to text analysis. But very curious what you were working on?
74.
by kianN 1y ago
I’ve actually run into the exact same issue. At the time we similarly had to scrap bandits. Since then I’ve had the opportunity to do a fair amount of research into hierarchical dirichelete processes in an unrelated field. On a random day,
75.
DuckDB Sparse Array Functions(github.com)
1 points by kianN 1y ago|0 comments
76.
by kianN 1y ago
Hi all, author here. I originally was just planning to put together a few helper functions to support sparse arrays in DuckDB, but there ended up being a lot of interesting lessons in the process of benchmarking storage and optimizing queri
77.
Reducing Data Lake Storage by 52% with SciPy-Style Sparse Arrays in DuckDB(blog.sturdystatistics.com)
2 points by kianN 1y ago|1 comments
78.
by kianN 1y ago
My applications have focused on noisy, high dimensional small datasets in which it is either very expensive or impossible to get more data. One example is rare class prediction on long form text data eg phone calls, podcasts, transcripts. O
79.
by kianN 1y ago
Totally agree and great point that hierarchical models have been around for a long time; however, these were primarily analytical, leveraging conjugate priors or requiring pretty extensive integration. I would say his work with Stan and his
80.
by kianN 1y ago
I might checkout Statistical Rethinking given how frequently it is being recommended! Edit: Haha I just found the textbook and I’m remembering now that I actually worked through sections of it back when I was working through BDA several yea
81.
by kianN 1y ago
I don’t mean for the bar to sound too high. I think working through khan academy’s full probability, calculus and linear algebra courses would give you a strong foundation. I worked through this book having just completed the equivalent cou
82.
by kianN 1y ago
This book is very relevant to those fields. There is a common choice in statistics to either stratify or aggregate your dataset. There is an example in his book discussing efficacy trials across seven hospitals. If you stratify the data, yo
83.
by kianN 1y ago
This is my favorite book on statistics. Full stop. The author Andrew Gelman created a whole new branch of Bayesian statistics with both his theoretical work on hierarchical modeling while also publishing Stan to enable practical application
84.
by kianN 1y ago
It’s even worse with parlays: the events are potentially negatively correlated. There may be a 70% chance that Giannis has 3 or more turnovers and a 70% chance the bucks win, but the odds of both happening are less than 49% because more tu
85.
by kianN 1y ago
This is a great summary of why despite so much progress/tricks being discovered, so little progress to the core limitations to LLMs are made.
86.
by kianN 1y ago
These issues are often attributed to a bad implementation of AI, but I think the problem is a little more fundamental. The potential of AI that causes VCs and investors to swap their eyes for dollar signs is the ability to take unstructured
87.
Coursera's Monetization Journey: From Zero to IPO(classcentral.com)
1 points by kianN 1y ago|0 comments
88.
by kianN 1y ago
I think the children's programming is a really undiscussed aspect of this. Some investments don't have immediately measurable outcomes. But as someone whose parents worked long hours growing up, I'm really grateful that my ex
89.
by kianN 1y ago
I built a public tool a while back for some of my friends in grad school to support this sort of deep academic research use case. Sharing in case it is helpful: https://sturdystatistics.com/deepdive?search_type=external&
90.
by kianN 1y ago
This is exactly the challenge. When embedding were first popularized in word to vec they were interpretable because the word2vec model was revealed to be a batched matrix factorization [1]. LLM embedding are so abstract and far removed from

More ›