tanichu read “Philosophizing Statistics” https://amzn.to/3xP41RG Jun Otsuka @junotk_jp ） Read. This was a very good book!
It is a book of “philosophy,” but it starts with a basic discussion of statistics in science, and while describing without skipping (or even vaguely talking about) distribution formulas, tests, Bayesian inference, and so on, it develops its semantics and epistemology …, that is, what it means for our (truth) quest and our perception of the world, while situating it in the historical context of the philosophy of science. The book is a very interesting and interesting way to learn about the history of the philosophy of science, and to understand what it means for our search for (truth) and our perception of the world. A book that “properly straddles/connects” literature and science.
I have tended to think that discussions of the philosophy of science, which include statistics in the workings of science, have been somewhat inaccessible to me, and that everyone tends to end up discussing the world from a propositional-logical viewpoint.
The presentation was easy to understand, including a clear discussion of p-hacking and the philosophical background behind the creation of hypothesis testing. This book could be used as a textbook in introductory education at universities/research institutes.
I was curious about it, but it was loaded, and Mr. Maruyama ( @rmaruy I read this book because it was highly recommended by a friend of mine (Mr. Maruyama, thank you very much!). (I know it’s not a biblio-battle, but human recommendation is important.)
By the way, here’s a bit of my own story: …
Self-Story 1) Scientific Discovery and Collective Predictive Coding
Since around the end of last year, I have been discussing my own “collective predictive coding” = “symbolic emergence” argument (https://x.com/tanichu/status/1768431204196569470…)) to connect it to an overarching explanatory principle that describes the entire system of scientific > inquiry. I’m discussing a statement that explicitly connects the statement to autonomous scientific discovery by AI, (I’ve also made a few grazing remarks at the CRDS AI x Philosophy WS. https://jst.go.jp/crds/report/CRDS-FY2023-WR-03.html…)) In such a discussion, your talk about placing the activity of scientific discovery itself as statistical modeling in the context of philosophy of science seemed to me to be a very important piece, in a sense, to give concrete form to my current faltering discussion (I will discuss it with you in person soon). (I will be discussing this with you in person soon.)
My Story (2): The History of Otsuka Sensai and Myself
Actually, I realized the other day that Mr. Otsuka and I met about 17 years ago. After I finish my degree,
“How can a constructivist approach be part of science? I was in the midst of worrying that what I was doing was perhaps non-scientific? (I was in the middle of my worries about “Is what I am doing possibly nonscientific? I was away from home at that time.
(Incidentally, it took me about 5 years after that to settle that concern to a certain extent in my mind, “Symbolic Emergence Robotics” (2014) https://amzn.to/3JsJSTT (It is written in Chapter 6, “The Constructivist Approach.“)
At that time, it was long before the third AI boom, and the majority view was not focused on AI/robotics-based human understanding, but I remember having a (maybe) lunch with your comment about the relationship between model theory and simulation philosophy in science. We hit it off that day, but I had not been able to maintain such a collaboration since then because my research history was still on the eve of symbolic emergence robotics and I had not yet established my own research. (Actually, I don’t even remember your name properly, just a fluffy episodic memory.)
Just a few days ago, I was in a chat space in a certain place (https://x.com/rmaruy/status/1766751409884868712…)) and
“Mr. Otsuka, maybe you are that guy from that time, right …?” “Mr. Taniguchi, oh, I knew it. …!”
We had a mysterious reunion like that. (This is the funny thing about life: ～～～～～～～～)
I hope that those who have not read it will read it and think about the position of their own research and study, what they need to pay attention to in hypothesis generation and statistical processing, the impact of AI based on deep learning on our future scientific discoveries and perceptions, and the future “view of knowledge” and so on. I hope you will enjoy the presentation.
junotk_jp Thank you very much! I am very happy to hear you say so, and I am so glad that I wrote it. I am excited to be able to work with you again after 15 years…!
yosss66 Wow - such a book! I wish I had it when I was in the service! I’m not sure if I’ll need it anymore, but I’ll buy it…
This page is auto-translated from /nishio/統計学を哲学する using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.