More to provide another perspective, I'll give the citation of some work
with Harry Joe and myself from over 2 decades ago.
@Article{,
author = {Joe, Harry and Nash, John C.},
title = {Numerical optimization and surface estimation with imprecise
function evaluations},
journal =
Hi Bert,
The article notes that chatGPT often gets the concept wrong, rather
than the facts. I think this can be traced to the one who poses the
question. I have often encountered requests for help that did not ask
for what was really wanted. I was recently asked if I could
graphically concatenate
Thanks, Ben.
For certain reasons, I would *not* like to apply global optimization solvers,
e.g., for reasons of higher dimensions and longer running times.
I was hoping for suggestions from the "Stochastic Programming" side.
And please, never suggest `optim` with method "SANN".
See the
It does often behave better if you say to it "that doesn't seem to be
working" and perhaps some error message
It is afterall a language tool. Its function is to provide text that seems
real.
If you ask it a science question and ask it to provide references in
Vancouver format, it can format the
This is a huge topic.
Differential evolution (DEoptim package) would be one good starting
point; there is a simulated annealing method built into optim() (method
= "SANN") but it usually requires significant tuning.
Also genetic algorithms.
You could look at the NLopt list of
While working on 'random walk' applications, I got interested in
optimizing noisy objective functions. As an (artificial) example, the
following is the Rosenbrock function, where Gaussian noise of standard
deviation `sd = 0.01` is added to the function value.
fn <- function(x)
Thanks.
https://www.wsj.com/articles/with-ai-hackers-can-simply-talk-computers-into-misbehaving-ad488686?mod=hp_lead_pos10
Ever heard of AI prompt injection?
*Stephen Dawson, DSL*
/Executive Strategy Consultant/
Business & Technology
+1 (865) 804-3454
http://www.shdawson.com
On 8/13/23
**OFF TOPIC** but perhaps of interest to some on this list. I apologize in
advance to those who may be offended.
The byline:
"ChatGPT's odds of getting code questions correct are worse than a coin flip
But its suggestions are so annoyingly plausible"
8 matches
Mail list logo