Rank: 3742 based on 8 downloads (last 30 days) and 1 file submitted

Roland Pihlakas

E-mail: Contact Author
Company/University: Simplify (Macrotec LLC) / University of Tartu
Lat/Long: 58.35800170898438, 26.7189998626709

Personal Profile:

Homepages: http://www.simplify.ee/ and http://roland.pri.ee/

My interests: AGI, connecting right people with right information and opportunities, art of understanding, philosophy.
BSc in psychology. Now doctorate student, researching safety of AI.

Competence:
- Combinatorial optimisation algorithms
- Safe artificial intelligence
- Modelling of natural general intelligence
- Artificial intelligence
- Optimisation algorithms
- C++ preprocessor and template metaprogramming
- GPGPU programming / RapidMind, Brook+
- 3D Robot simulations
- Robotics

Research:
- Safety of artificial intelligence
- Modelling of natural thinking

Professional Interests:

artificial intelligence, learning algorithms, combinatorial optimisation, constraint programming

Watch this Author's files

Files
Watch Lists

Files Posted by Roland					1 - 1 of 1
Updated		File	Tags	Downloads (last 30 days)	Comments	Rating
07 Dec 2010		Methods for calculating precise logarithm of a sum and subtraction Methods for operating on log's of values without calculating the original values during the process Author: Roland Pihlakas	mathematics, statistics, article, logarithm, paper	8	1

Comments and Ratings by Roland		View all 1 - 2 of 2
Updated	File	Comments	Rating
29 Sep 2009	GODLIKE - A robust single-& multi-objective optimizer GODLIKE combines 4 global optimizers for both single/multi-objective optimizations Author: Rody Oldenhuis	A bugfix. set_options.m, at lines 456-457: elseif strcmpi(option, 'SkipTest') if ~isnumeric(value) should be: elseif strcmpi(option, 'SkipTest') if ~ischar(value)
02 Sep 2009	GODLIKE - A robust single-& multi-objective optimizer GODLIKE combines 4 global optimizers for both single/multi-objective optimizations Author: Rody Oldenhuis	Hi, You can make it orders of magnitude faster if You replace in pop_single.m this: for i = 1:pop.dimensions % convert column to decimal representation temp_pop(:, i) = sum(convert_to_dec.newpop(:, 1:NumBits), 2); % delete entries newpop(:, 1:NumBits) = []; end with this: newpop_startcol = 1; NumBits2 = NumBits-1; for i = 1:pop.dimensions % convert column to decimal representation temp_pop(:, i) = sum(convert_to_dec.newpop(:, newpop_startcol : newpop_startcol+NumBits2), 2); % delete entries newpop_startcol = newpop_startcol + NumBits; end Column/row deletion is very slow operation, its better to avoid it. There were also some changes I had to make in order for the script to start in Matlab R2007a. Are You considering to port it to mex file? I am going to do that but we may get the results sooner if we split the task.

Comments and Ratings on Roland's Files			1 - 1 of 1
Updated	File	Comment by	Comments	Rating
10 Sep 2009	Methods for calculating precise logarithm of a sum and subtraction Methods for operating on log's of values without calculating the original values during the process Author: Roland Pihlakas	Rowland, Darren	I can see this being a useful utility for some applications. However the code could be improved. I would like to see you add - a H1 line - an example (or two) showing usage, particularly comparing your method with a method which will produce overflow - Comments within the code explaining the procedure so a user does not need to consult the pdf file Lastly I think the actual computation should be on just one line i.e. R = log(exp(a_ln - b_ln) + 1) + b_ln; rather than spread out over many lines. Let comments explain the shifting. Darren

Top Tags Applied by Roland	1 - 5 of 5
article, logarithm, mathematics, paper, statistics

Files Tagged by Roland					1 - 1 of 1
Updated		File	Tags	Downloads (last 30 days)	Comments	Rating
07 Dec 2010		Methods for calculating precise logarithm of a sum and subtraction Methods for operating on log's of values without calculating the original values during the process Author: Roland Pihlakas	mathematics, statistics, article, logarithm, paper	8	1

Currently Watching
Type	Description
Author	Rody Oldenhuis
File	GODLIKE - A robust single-& multi-objective optimizer
File	Many testfunctions for global optimizers
File	Improving Optimization performance with Parallel Computing
File	Methods for calculating precise logarithm of a sum and subtraction
File	fminsearchbnd, fminsearchcon
File	Optimization Tips and Tricks
File	cprintf - display formatted colored text in the Command Window
File	Learning the Kalman Filter
File	Kalman Filter Tutorial
File	press

Files Matching Roland's Watch List				View all 1 - 5 of 21
Updated	File	Tags	Downloads (last 30 days)	Comments	Rating
06 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch Author: John D'Errico	optimization, fminsearch, bounds, constraints, nonlinear, inequalities	308	46	4.97368 5.0 \| 38 ratings
27 Nov 2011	cprintf - display formatted colored text in the Command Window Displays sprintf-formatted strings in the Command Window using the specified color/underline style Author: Yair Altman	java, desktop, gui, undocumented	130	29	4.89655 4.9 \| 29 ratings
11 May 2011	Optimization Tips and Tricks Tips and tricks for use of the optimization toolbox, linear and nonlinear regression. Author: John D'Errico	fminsearch	179	43	4.77273 4.8 \| 47 ratings
19 Feb 2011	RKN1210 - A 12th/10th order Runge-Kutta-Nystrom integrator Integrator for second-order ODE's with very stringent error tolerances. Author: Rody Oldenhuis	accuracy, secondorder ode, rungekuttanystrom, highorder, numerical integration, ode	14	0
07 Dec 2010	Methods for calculating precise logarithm of a sum and subtraction Methods for operating on log's of values without calculating the original values during the process Author: Roland Pihlakas	mathematics, statistics, article, logarithm, paper	8	1

Comments and Ratings Matching Roland's Watch List			1-5 of 322
Date	File	Comment by	Comments	Rating
06 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch	John D'Errico	(I've fixed the bug about the outputfcn problem and re-posted it. The new version should appear sometime this morning.) As far as Stefan's problem goes, anytime you throw problems with variables that vary by 20 to 30 powers of 10 at ANY numerical optimizer, expect serious things to go wrong. This is a no-no for literally ANY numerical tool. And setting a convergence tolerance (TolX) to eps is just as silly, especially when your parameters vary by that many orders of magnitude. Essentially, you are asking that one variable, which can apparently vary somewhere between 1 and 1e30, must be computed to an absolute accuracy of roughly 2e-16. A common solution is to normalize your variables, so that both are at least of similar orders of magnitude. If you were trying to solve a problem where one variable had bounds of 1e20 to 5e20, and a second variable was bounded in the interval 1e-6 to 1e-5, then normalizing the variables to both be of order 1 would make complete sense. But your variables each vary by many powers of 10! One thing all novices need to learn about computing, is that when things vary by that many powers of 10, is to use logs! Let fminsearch vary the log10 of your variables, so they now have bound vectors that look like [0, -6] for the lower bounds, and [30, 0] for the upper bound. Now, inside your function, take the parameters and raise 10 to that power BEFORE they are used. Do the same with the output when it is returned. As far as fminsearch is concerned, your numbers are perfectly normal looking, but your objective sees the numbers in their proper scales. You will find that any optimizer runs much more happily when you do this, as now the variables are very nicely normalized. All of the mathematics works more simply when you work in the log space. In fact, this is surely the natural way to work with variables that vary by many powers of 10. A nice consequence is the tolerances in TolX become relative tolerances on the variables, something that surely makes much more sense. And DON'T use eps as a convergence tolerance. Fminsearch will never be able to give you even a reasonable shot at convergence in even 10000 function evals with that fine of a tolerance. So what happens is fminsearch will keep on iterating until it runs out of function evaluations. Use a meaningful, realistic tolerance. You were probably only setting that fine of a tolerance because of the ridiculous variable scaling anyway.	Comment only
06 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch	Stefan	The following test shows strange results: ---------------------------------- G0bnd 2.22e+014 L0bnd 2.707e-006 G0 2e+014 L0 3e-006 ---------------------------------- function test() options=optimset('Display','on', ... 'MaxFunEvals',1e4, ... 'MaxIter',1e4, ... 'TolFun', eps, ... 'TolX', eps, ... 'FunValCheck', 'on' ... ); w= 150e-6; xData = 1e-6:1e-6:w; G0 = 2e14; L0 = 3e-6; yData = G0exp(-xData./L0); start_params = [G0, L0]; min_params = [ 1, 1e-6]; max_params = [1e30, 1]; plot(xData,yData,'-r'); hold('on'); result_params=fminsearchbnd(@fitG_error_function,start_params,min_params,max_params,options,xData,yData); ['G0bnd ' num2str(result_params(1),4) ' L0bnd ' num2str(result_params(2),4)] plot(xData, result_params(1)exp(-xData./result_params(2)),'-g'); result_params=fminsearch(@fitG_error_function,start_params,options,xData,yData); ['G0 ' num2str(result_params(1),4) ' L0 ' num2str(result_params(2),4)] plot(xData, result_params(1)exp(-xData./result_params(2)),'-b'); function fiterror = fitG_error_function(start_params,xData,yData) Fitted_Curve= start_params(1) exp(-xData./start_params(2)); Error_Vector=Fitted_Curve - yData; fiterror=sum(Error_Vector.^2); results.fiterror=fiterror; end end	Comment only
06 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch	Romain	Great function, very handy. I have noticed a small error, though. If fminsearch is not called, the function returns output.funcount, whereas it returns output.funcCount otherwise.	5
01 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch	John D'Errico	Sorry. I'd been playing with various alternatives sometime ago. The name on the header will be correct now when the latest update comes alive.	Comment only
01 Feb 2012	fminsearchbnd, fminsearchcon Bound constrained optimization using fminsearch	Martin Richard	John, I've been using that function for years now. Love it. Perhaps you may want to change the function name so that it fits the file name in the new version (fminsearchbnd3 in new file).	5