John Carlin says, “‘Identifying variables that independently predict…’ is not a well-defined research task”

2026-05-02
John “Bayesian Data Analysis” Carlin writes: Recent developments in the methodology of epidemiological research have emphasized the importance of achieving clarity of purpose by classifying research questions into one of three types: descriptive, predictive, and causal… . I … Continue reading → …

Bad Weather and the Subway

2026-05-02

Snow in Inwood, New York. Photograph by the author.

Recently I’ve been looking at hourly ridership data from the New York City Subway. Last time we learned that people go to work in the morning and come home in the eve…

Continue reading: Bad Weather and the …

You Don’t Need to Learn All the Weights on tabular data: The Case for rvflnet (a nonlinear expressive glmnet) on regression, classification and survival analysis

2026-05-02
rvflnet is an R package that implements a Random Vector Functional Link (RVFL) network. It is a nonlinear expressive version of glmnet that can be used for regression, classification and survival analysis.
Continue reading: <a href=“https://www.r-bloggers.com/2026/05/you-dont-need-to-learn-all-the-weights-on-tabular-data-the-case-for-rvflnet-a-nonlinear-expressive-glmnet-on-regression-classification-and-surviv …

一道医疗背景的笔试题

2026-05-02

本地科研助理

2026-05-02

去年年底我就开始关注龙虾类应用，自己也折腾了一段时间，热度过去基本就闲置了，更多时候还是用自然语言编程。后来仔细回忆了一下，我其实不排斥搞一个个人助理，真正没放开用是因为不了解其实现过程，担心出错跟信息泄露。那解决方案就很简单了，自己搭一个。

我的核心诉求很简单，就是要一个能在 Telegram 和 QQ 里对话的科研助理。这里面有几个要素需要保证：