Kaizen-R.com new home

Thinking about Systems

complexity

ABM

systems

I did mention some of it a while back, but also I am re-reading “Thinking in Systems” by D. H. Meadows.

RLCS with the mirai package

RLCS

code

I have actually now have had the time to evaluate (only briefly) the use of the mirai R package and it’s… interesting.

RLCS: Less sequential?

RLCS

code

I’ve let time pass lately as I looked into other topics (ODE and migrating content from the old Blog).

Nonlinear Dynamics: on Bifurcations

math

code

complexity

It turns out, back in February 2022, I published on the “old” Blog an entry on the Logistic Map.

Studying Nonlinear Dynamics: Day 1

math

code

complexity

Math. More specifically, Ordinary Differential Equations. More specifically, in fact, probably things related with evolution in time. Although not only.

RLCS package: Progressing nicely

RLCS

Yesterday was the first day of testing.

RLCS as a package?

RLCS

I have plenty of work still, with this RLCS Package idea.

RLCS for text? And a new idea…

NLP

ML

RLCS

Alright so this a bit of a crazy idea.

Morning Coffee Thoughts

Decisions

Just a quick note to myself here on a realization.

Classic Computational Social Model: Schelling’s Segregation

Decisions

Optimization

visualization

complexity

And I thought some of these ideas were quite interesting. I don’t know yet what to make of it all (I haven’t read it all, either). It feels some of it is related to…

Multi-Agent RL

RLCS

ML

Really all I did here, is modify the simple World I created a few months back to support multiple agents (blue-dots).

RLCS: Caring for better performance

RLCS

code

Optimization

There was still room for improvement. Probably still is, of course, just… Not with a total redo, I guess.

RLCS: Making things faster where it matters

RLCS

code

Optimization

This entry was not “planned”, as I am still working on this last bit, and also I shouldn’t touch things as I have a presentation coming next week on this, but…

RLCS: small code rewrites

ML

RLCS

code

Optimization

As announced, I’m doing some code rewriting, just because…

RLCS: Better code?

ML

RLCS

code

Optimization

Most of the code thus far was about getting it to work, and for it to not be too slow at runtime (with several approaches considered).

Sunday Thoughts: About Strategy, Information, Decisions, Simulations, Reinforcement Learning & al.

Information

ML

Decisions

Optimization

Yesterday I did nothing. Like… Nothing. I considered it an active choice, and as such I wasn’t sad at the end of the day for not having “produced” anything… And today I was…

RLCS: Small addition

ML

RLCS

I was in attending a local R conference the other day, and took the opportunity to discuss a bit with some of the (very few) that have a minimal interest (and an…

RLCS for Data Mining: Visuals

ML

RLCS

visualization

I’ve shown quite a few “Classifiers” and detailed how data mining can be cool with an LCS.

RLCS & Explainable AI

ML

RLCS

visualization

I have been open about some of my worries with GenAI (and in fact Deep Learning with Neural Networks).

RLCS for RL: It works again (using books helps)

ML

RLCS

Recently, I had “broken” my Reinforcement Learning implementation, in the sense that it would learn at first, and then it would get worse with more experience, instead of…

Migrating Contents (slowly)

news

I’ve had my own kaizen-r.com blog for some time, and I relaunched on github.io in November last year.

RLCS: Improving… And getting worse (at the same time!)

ML

RLCS

Well, it’s been a couple of weeks (I was a bit offline last couple of weekends, I guess). So I’m back at it and, first, I needed to clean things up. It just was messy code.

RLCS: Reinforcement Learning: World Explorer v1

ML

RLCS

This is quite literally my first Reinforcement Learning exercise ever. Coded from scratch. With no Neural Network whatsoever, but instead, using my own implementation of the…

RLCS: A World to Play RL

ML

RLCS

Today I just started something simple: An Object to represent a simulated World in which an Agent can move.

RLCS: Reinforcement Learning?

ML

RLCS

At least the XCS version of LCS algorithm is a “well-known” (in the rather small sub-world of LCS enthusiasts, that is) approach to do Reinforcement Learning.

RLCS: training in parallel?

ML

RLCS

I’ve already tried to parallelize a bit the training of a supervised-learning LCS. It kinda’ works… But the more I think about it, the more I’m convinced it’s not as easy as…

RLCS: for Data Mining

ML

RLCS

The exact same code to do supervised learning can be used for data mining. In fact, I’ve already shown that, with the “not-bit-4” example.

RLCS Day 4.5 - Fully Functional v0 and New Tests

ML

RLCS

Version “0” is done and working. And I will say this, it is already showing some of its power. Let me show you.

RLCS Day 2 - Full Rule Discovery

ML

RLCS

I have yet another day off (that’s the thing with Holiday season), and I really wanted to keep working on this.

RLCS Day 1 - Part II: Basic Covering

ML

RLCS

So I had a bit of time. And I chose to use it to keep going a bit. I moved a couple of things, created a few more supporting functions, modified them somewhat, and added the…

RLCS Day 1: Rules formatting, storage, and matching

ML

RLCS

So I wasn’t kidding. I really like this “Learning Classifier System” (aka LCS) idea, and I really want to develop the R package for it, which I will call “RLCS” (clever, hu?).

A new project

ML

news

RLCS

Ever since I read about the concept in M. Mitchell’s book “Complexity The emerging science at the edge of order and chaos” some time last year (I think around this time of…

Understanding Linear Regression

ML

math

This is probably one of the last entries from my preparation to introduce “Background of ML for Cybersecurity”. This time, it’s really about what should probably have been…

Classifying URLs

ML

code

I’m still working my way through a “practical demos” session on ML background for Cybersecurity.

Entropy - Identifying compressed files

math

code

About Shannon’s Information Entropy, applied to potentially detecting ciphered or compressed text compared to plain text.

Welcome To My Blog

news

This is the first post in this future Blog. Currently testing all kinds of things, will updated hopefully shortly!

ML Concepts: Word Embeddings

ML

NLP

I discussed an unsupervised clustering algorithm, DBScan. We used multi-dimensional points (coordinates) in space. But what if our data is text? That’s common in “general”…

ML Concepts: Unsupervised Learning – Clustering – DBScan

ML

visualization

So as I keep working on an upcoming presentation I shall give in a few weeks, I thought I’d prepare something a bit more “live” than just a PPT with Memes (although who…

Random Exercises: Interpolation with Lagrange Polynomials

math

As I want to make sure I keep posting here, I keep looking for things to write about in my own personal “library” (book shelves).

Testing LLMs locally on basic Apple Silicon

ML

I have learned since. But back in October 2024, I did run the following test.

Cholesky’s matrix decomposition and where not to use it

math

As I was asked to talk about AI for Cybersecurity, I have been thinking about it for a few weeks, refreshing concepts, taking some notes, gathering some ideas, sources…

MSC Thesis delivered

math

In the old blog, I did comment/discuss a lot about the latest Master Thesis I was working on.

While on the Train: Cellular Automata

visualization

I decidedly LOVE riding on a train: For whatever reason, it’s one particular moment where I somehow focus a bit more than usual. It’s usually just a few hours, reasonably…

Interlude

complexity

But then… I finished this book, and I liked it, and I thought I would recommend it.

Optimization (4/n): Genetic Algorithm(s) (2/2)

Optimization

This week, I’ll make it short, and instead of boring with code and explanations, I thought I’d just show an example output…

Optimization (4/n): Genetic Algorithm(s) (1/m)

Optimization

This time around, I keep going studying and implementing optimization algorithms, based on the already mentioned reference book, and I want to work on a slightly different…

Optimization (3/n): Simulated Annealing

Optimization

Continuing with this simple “series” (see here and here), I implement the next algorithm proposed by the reference book, but in R.

Optimization (2/n): Gradient Descent

Optimization

math

So I keep going with the reference book for a while. BTW, the book can be found here (NOT a referral or anything, I don’t make money out of this, it’s just a pointer to one…

Summer fun: Testing Optimization Algorithms (1/n)

Optimization

I have some more spare time for the upcoming few weeks, and I haven’t had the opportunity to take the “Metaheuristics” course in the Master, and so I feel I’m missing out a…

A quick look at CISA KEV

cyber

I keep hearing about it, so it was time I had a look at that famous CISA “Known Exploited Vulnerability” dataset. Also, it’s been a while since I did something more directly…

First Entry: Variables

code

This one was my first-ever Blog post on the original kaizen-r.com site. I’m just reproducing it here out of… Nostalgia, I guess.