Yi-Shin Lin

I gave a talk on Utility Maximization Model, visit osf.io/av3cm/ for the presentation slide and more.

@hcp4715 @crblandet "the neural basis of cognitive functions, diagnosing and treating brain disorders, and brain-inspired computing." These three goals seem to have clear enough scientific significance in term of the scale of the project. For example, a GPU based software may do the third?

@crblandet I am not an expert on this. Rao Yi's opinion, from his blog, is that the primate models related to Bmal1 from Poo's team do not have clear scientific or practical significance ("科学和应用意义都不明确（或者说都比较差）的研究")

science.org/content/articl… Rao Yi criticized the lack of transparency of the finance behind this big project, on his wechat account (mp.weixin.qq.com/s/BmLWL0zKrsW7…) Will the criticism help? I am very pessimistic.

免費參加里茲大學人因工程研究群的數學建模方法讀書會github.com/yxlin/Computat…

Join our discussion on computational modelling. See more details at github.com/yxlin/Computat…

"I saw that even ppl w/ a lot of power felt their hands are tied. It really puts into perspective how brave & how much risk ECRs are taking when they stand up to the system. I wanted to help provide a community for ppl who are taking those risks bc I think it's so admirable" -me

arxiv.org/abs/2110.11015 We develop a computational model basing on three simple cognitive principles, maximizing utility, discrete motor primitive, and sensory evidence accumulation to account for road user interactions. Find software and details in the preprint.

This is now published. If you're interested in how people use, and misuse, automated advice you should read it 🦾💪. We present a model of the underlying cognitive processes involved. journals.sagepub.com/doi/10.1177/09…

Enjoy the powerful statistical tools in R? Want to apply them in analysing EEG data. Matt Craddock's eegUtils is abwonderful R package that give the user the best of both worlds.

The Practical Alternative to the p Value is a System Encouraging Good Scientific Practice

What’s the purpose of mathematical modeling? statmodeling.stat.columbia.edu/2021/06/23/452…

Data visualisation using R, for researchers who don't use R zpr.io/R5hZs

facetscale, flexibly allow each facet panels have their own scale limit and tick mark. github.com/zeehio/facetsc…

Doing signal processing, pracma and signal are quite useful, cran.r-project.org/web/packages/s…

Struggling to find distinguishable and aesthetic colours to differentiate different conditions? Love the NZ birds github.com/G-Thomson/Manu

A fantastic plotting tool cran.r-project.org/web/packages/c…

Ever get that feeling your model may be either severely ill-conditioned or misspecified? Been one of those days. If you're head-desking, know you're not alone.

Watch for your prior. Uninformative is also a belief.

Anticipation responses , RT < 50 ms, defeat smart samplers.

Sequential Bayes Factor designs in developmental research: Studies on early word learning - very exciting new paper by @nivedita_mani and colleagues showing how we can maximize the efficiency of developmental work (plus three nice replication studies!) onlinelibrary.wiley.com/doi/full/10.11…

My rebuttal #rdatatable