Arto Bendiken artob

Accelerating Discrete Program Search

I am investigating how to use Bend (a parallel language) to accelerate Symbolic AI; in special, Discrete Program Search. Basically, think of it as an alternative to LLMs, GPTs, NNs, that is also capable of generating code, but by entirely different means. This kind of approach was never scaled with mass compute before - it wasn't possible! - but Bend changes this. So, my idea was to do it, and see where it goes.

Now, while I was implementing some candidate algorithms on Bend, I realized that, rather than mass parallelism, I could use an entirely different mechanism to speed things up: SUP Nodes. Basically, it is a feature that Bend inherited from its underlying model ("Interaction Combinators") that, in simple terms, allows us to combine multiple functions into a single superposed one, and apply them all to an argument "at the same time". In short, it allows us to call N functions at a fraction of the expected cost. Or, in simple terms: why parallelize when we can sha

This is the predecessor of Mazeppa.

Supercompilation is a deep program transformation technique due to V. F. Turchin, a prominent computer scientist, cybernetician, physicist, and Soviet dissident. He described the concept as follows [^supercompiler-concept]:

A supercompiler is a program transformer of a certain type. The usual way of thinking about program transformation is in terms of some set of rules which preserve the functional meaning of the program, and a step-by-step application of these rules to the initial program. ... The concept of a supercompiler is a product of cybernetic thinking. A program is seen as a machine. To make sense of it, one must observe its operation. So a supercompiler does not transform the program by steps; it controls and observes (SUPERvises) the running of the machine that is represented by th

Personal notes building zig 0.14.0-dev.xxxx on Macbook M3 Max

This probably works on all current Apple Silicon Macs, but you must check other models yourself. Newer Intel Macs are also likely to work, with one significant difference: homebrew has a different default path.

20230926: Updated for macOS Sonoma, latest XCode Commandline Tools, homebrew-installed LLVM 17.01

20231001: Build zig binaries for both release and debug mode (with release mode as default)

We need to talk about group.

I'm sorry to tell you, but group is gone.

If you check out the latest k.d on shakti.com as at 28 March 2023, you'll see that 'unary' = is now 'freq' (frequency).

Group had a long life: it's been around since at least k2, or 1996.

So why did group go? And what should we use instead?

The reverse turing test

I asked GPT-4 to come up with 10 questions to determine if the answerer was AI or human.

I provided my own answers for these questions and I also asked ChatGPT to answer them.

The result is that GPT-4 was able to correctly differentiate between AI and Human.

Purpose

Bootstrap knowledge of LLMs ASAP. With a bias/focus to GPT.

Avoid being a link dump. Try to provide only valuable well tuned information.

Prelude

Neural network links before starting with transformers.

Some remarks on Large Language Models

Yoav Goldberg, January 2023

Audience: I assume you heard of chatGPT, maybe played with it a little, and was imressed by it (or tried very hard not to be). And that you also heard that it is "a large language model". And maybe that it "solved natural language understanding". Here is a short personal perspective of my thoughts of this (and similar) models, and where we stand with respect to language understanding.

Intro

Around 2014-2017, right within the rise of neural-network based methods for NLP, I was giving a semi-academic-semi-popsci lecture, revolving around the story that achieving perfect language modeling is equivalent to being as intelligent as a human. Somewhere around the same time I was also asked in an academic panel "what would you do if you were given infinite compute and no need to worry about labour costs" to which I cockily responded "I would train a really huge language model, just to show that it doesn't solve everything!". We

Safari's date-picker is the cause of 1/3 of our customer support issues

...and obviously we're building a workaround. But I'm absolutely flabbergasted that a standard <input type="date"> HTML field, in a standard browser, from a company that bases its reputation good design, could be so dreadful.

The context

I'm the developer for a startup that sells a genetic test to recommend medications for high blood pressure. For medical reasons we need to know our customers' birth date. Most of our customers are in their 60s or older. We've found that many of them use iPads or iPhones. And they're the ones who complain to our customer support that our site is unusable.

The problem

ChatGPT passes the 2022 AP Computer Science A free response section

For fun, I had ChatGPT take the free response section of the 2022 AP Computer Science A exam. (The exam also has a multiple-choice section, but the College Board doesn't publish this.) It scored 32/36.

Methodology

For each question, I pasted in the full text of the question and took the response given.
I tried each question once and took the response given: no cherry-picking. For readability, I've added indentation in some cases, and included method signatures where they were provided in the question and ChatGPT only provided a body. I've added question numbers; any other comments are ChatGPT's.
Many questions have examples containing tables or diagrams; because those don't translate well to plain text, I excluded those tables/diagrams and any text that referenced them.
I excluded the initial instructions at the top of th

	import 'dart:ffi';
	import 'dart:isolate';
	import 'dart:io';
	import 'package:ffi/ffi.dart';

	void newIsolate(SendPort sendPort) {
	final ptr = calloc.allocate<Int>(100);
	ptr[0] = 42;
	sendPort.send(ptr.address);
	sleep(const Duration(seconds: 1));