**statistics**

[WIP] Giant refactor!!!!!!!!! by apg · Pull Request #5 · heroku/barnes

1 hour ago by dentarg

@amerine one thing that trashed does that's sort of strange is assign an arbitrary sample rate to the per request data. So, I'm not convinced it's very accurate, and in order to support it, there's quite a lot of complexity in the code that as an outsider (Ruby is by far not my first language) I had trouble piecing together.

Heroku metrics doesn't currently have any understanding of per request metrics (we aggregate 5s of requests into a histogram for response times before they even hit Kafka) and so per request metrics just don't currently fit in our model. As we get more sophisticated, I think there may be some opportunity for us to do sampling of request metrics that align with full HTTP requests. One could imagine that we instrument all requests whose request id ends in '0', or something silly like that in order to provide much greater visibility.

heroku
statistics
metrics
Heroku metrics doesn't currently have any understanding of per request metrics (we aggregate 5s of requests into a histogram for response times before they even hit Kafka) and so per request metrics just don't currently fit in our model. As we get more sophisticated, I think there may be some opportunity for us to do sampling of request metrics that align with full HTTP requests. One could imagine that we instrument all requests whose request id ends in '0', or something silly like that in order to provide much greater visibility.

1 hour ago by dentarg

The Algorithmic Foundations of Adaptive Data Analysis – Fall 2017, Taught at Penn and BU

5 hours ago by arsyed

"This class will take a mathematically rigorous approach to understanding how to mitigate overfitting and false discovery when doing data analysis in the common case in which data is repeatedly re-used, both to suggest which analyses should be performed, and to actually conduct those analyses."

courses
statistics
adaptive
5 hours ago by arsyed

[1609.06840] Exact Sampling from Determinantal Point Processes

18 hours ago by csantos

Determinantal point processes (DPPs) are an important concept in random matrix theory and combinatorics. They have also recently attracted interest in the study of numerical methods for machine learning, as they offer an elegant "missing link" between independent Monte Carlo sampling and deterministic evaluation on regular grids, applicable to a general set of spaces. This is helpful whenever an algorithm explores to reduce uncertainty, such as in active learning, Bayesian optimization, reinforcement learning, and marginalization in graphical models. To draw samples from a DPP in practice, existing literature focuses on approximate schemes of low cost, or comparably inefficient exact algorithms like rejection sampling. We point out that, for many settings of relevance to machine learning, it is also possible to draw exact samples from DPPs on continuous domains. We start from an intuitive example on the real line, which is then generalized to multivariate real vector spaces. We also compare to previously studied approximations, showing that exact sampling, despite higher cost, can be preferable where precision is needed.

sampling
Statistics
MachineLearning
Probability
18 hours ago by csantos

[1707.04345] Gaussian Graphical Models: An Algebraic and Geometric Perspective

18 hours ago by csantos

Gaussian graphical models are used throughout the natural sciences, social sciences, and economics to model the statistical relationships between variables of interest in the form of a graph. We here provide a pedagogic introduction to Gaussian graphical models and review recent results on maximum likelihood estimation for such models. Throughout, we highlight the rich algebraic and geometric properties of Gaussian graphical models and explain how these properties relate to convex optimization and ultimately result in insights on the existence of the maximum likelihood estimator (MLE) and algorithms for computing the MLE.

via:arthegall
GraphicalModels
statistics
AlgebraicGeometry
18 hours ago by csantos

Google - Talk to Books

22 hours ago by davesurgan

Browse passages from books using experimental AI

ai
book
Search
statistics
22 hours ago by davesurgan

NBA has baller season attendance, ratings, merchandise see huge uptick

22 hours ago by davesurgan

It's not just attendance seeing gains. TV ratings are surging as well. TNT's live NBA game telecasts averaged 1.7 million viewers this year, making it the most-watched regular season since 2013-2014. More importantly to advertisers, the coveted 18-34 and 18-49 demographic saw a 14 percent and 15 percent increase.

The NBA on ABC was up 17 percent, also experiencing double digit increases in several key demos.

This comes after the NFL and NHL both saw declines in TV ratings this season.

statistics
nba
basketball
The NBA on ABC was up 17 percent, also experiencing double digit increases in several key demos.

This comes after the NFL and NHL both saw declines in TV ratings this season.

22 hours ago by davesurgan