Xah Talk Show 2025-06-02 Ep662 Wolfram Language Explore Machine Learning Features

• https://youtu.be/eQbRGuD4mdo

• timestamp
• 5:59 Stephen Wolfram on ChatGPT
• 8:53 grok voice chat demo
• 13:23 twitter spam bots
• 14:36 Wolfram Alpha demo
• 18:03 Neural Network Universal Approximation Theorem
• 21:45 grok write tutorial on kotlin
• 27:58 Wolfram Alpha in Mathematica
• 34:02 7k function builtin Wolfram language
• 36:55 Wolfram language doc intro
• 44:12 cryptography demo
• 52:48 graph layout demo
• 53:37 plot 3d demo
• 55:11 machine learning
• 59:00 machine learning classify demo
• 01:16:32 command line repl vs notebook

• Wolfram Language Tutorial
• Emacs: Xah Wolfram Mode 📦

• wolfram alpha, how it diff from search engine, and ai chat bots
• the rise of ai chat bots, e.g. chatgpt
• Stephen Wolfram on ChatGPT
• demo grok voice chat
• Neural Network Tutorial
• Neural Network and The Universal Approximation Theorem

• grok wrote a simple tutorial on kotlin programing language
• https://x.com/i/grok/share/NBlhBPVG2cyMgyuXDUESwoJxi

• Command Line Interface (REPL) vs Notebook Interface

(* demo the classify machine learning function *)

(* generate a list of random numbers between 0 and 10 *)
xRandomList = Table[ RandomInteger[{0,10}] , {10}]
(* {8, 0, 9, 1, 9, 9, 5, 3, 8, 10} *)

(* now create a list of pairs, if number is less than 5, lable it small, if it is greater or equal than 5, label it big. *)
xData = Map[ Function[x, If[ x < 5, x -> "small", x -> "big"]] , xRandomList]
(* {1 -> small, 10 -> big, 1 -> small, 7 -> big, 7 -> big, 0 -> small, 9 -> big, 1 -> small, 7 -> big, 6 -> big} *)

xclassifyFun = Classify[xData];

xclassifyFun[3]

(* 
 | Time elapsed: | Training example used: | Current best method: | Current accuracy: | Current loss: |
 |     3.7s      |          8/10          |  LogisticRegression  |       0.833       |     0.235     |
small
 *)