You can subscribe to this list here.
2007 |
Jan
|
Feb
|
Mar
|
Apr
|
May
(5) |
Jun
(7) |
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
---|
From: Shu W. <sh...@gm...> - 2007-06-23 16:03:45
|
Over the past two weeks, I have some basic UI behaviors and object patterns in place for the UI. After telephone discussion with Bryan, we prioritized features in a UI spec came up with the following schedule: *Now - July 22th* work on the UI in a part-time capacity while focusing on finishing other projects *July 22 - Aug 4th* work full-time and closely with Bryan to deliver a working beta accomplishing must-have features in the UI Spec. ** *Aug 4 - Sept 5th* Work full-time to polish the beta and add in bells and whistles (ie. brain view). |
From: Bryan T. <bp...@gm...> - 2007-06-23 14:51:25
|
Last 2 weeks: Not much work on the simulator. A few bug fixes. Next 2 weeks: Not much again (I'll be busy with other things). After that I'll be working with Shu to make API changes needed for the UI. Bryan |
From: Chris E. <cel...@uw...> - 2007-06-18 14:11:09
|
Hi Shu, I'm quite familiar with Jeff's work, am good friends with the director of RNI, and have seen talks/read papers on it. Not to burst any bubbles, but from everything I've seen it looks more like a marketing exercise than interesting science. Hawkins' work hasn't contributed to state of the art theoretical neuroscience at all (in fact, he's rehashed old ideas and tried to sell them through his company). What he has tried to sell as a 'new theory' is the (old) observeration that there is significant repetitive structure (cortical microcircuits) to cortex (which is less than half the neurons in the brain) and that thus there should be some general computational principles explaining cortical function. The 'theory' is a poor implementation of Bayesian inference in a network which has never demonstrated good pattern recognition on any standard benchmarks -- they have a few results on non-standard recognition problems, which aren't challenging as far as I can tell. I hope you don't take this as a reason not to send me links you think are cool -- please do! Just thought you might want another opinion on what you found :) best, .c Shu Wu wrote: > Found this talk (from TED 2003) very interesting: > http://www.ted.com/index.php/talks/view/id/125 > > "To date, there hasn't been an overarching theory of how the human brain > really works, Jeff Hawkins argues in this compelling talk. That's > because we still haven't defined intelligence accurately. But one > thing's for sure, he says: The brain isn't like a powerful computer > processor. It's more like a memory system that records everything we > experience and helps us predict, intelligently, what will happen next. > Bringing this new brain science to computer devices will enable powerful > new applications -- and it will happen sooner than you think." > > > A link to the Redwood neuroscience institute: > http://www.rni.org/mission.html > > > -- > Shu Wu > http://www.sgsolutions.ca > > > ------------------------------------------------------------------------ > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > > > ------------------------------------------------------------------------ > > _______________________________________________ > Nesim-java mailing list > Nes...@li... > https://lists.sourceforge.net/lists/listinfo/nesim-java |
From: Shu W. <sh...@gm...> - 2007-06-16 10:04:08
|
Found this talk (from TED 2003) very interesting: http://www.ted.com/index.php/talks/view/id/125 "To date, there hasn't been an overarching theory of how the human brain really works, Jeff Hawkins argues in this compelling talk. That's because we still haven't defined intelligence accurately. But one thing's for sure, he says: The brain isn't like a powerful computer processor. It's more like a memory system that records everything we experience and helps us predict, intelligently, what will happen next. Bringing this new brain science to computer devices will enable powerful new applications -- and it will happen sooner than you think." A link to the Redwood neuroscience institute: http://www.rni.org/mission.html -- Shu Wu http://www.sgsolutions.ca |
From: Shu W. <sh...@gm...> - 2007-06-04 15:51:25
|
Past 2 weeks: - Developed some patterns for representing Nesim nodes as UI objects. Which should should translate into saved time later. - Basic UI Stuff: Drag and drop canvas, context menus, hover over behaviors - Was a little busy with work at MSRA, so spent a little less time on Nesim than I wanted Next 2 weeks: - Develop code for visual programming - Connecting nodes together, adding and removing connections via drag and drop, etc.. - should be much more productive with incorporating Nesim objects into the UI once the basic pieces are in place -Shu On 6/3/07, Hussein Hirjee <hah...@uw...> wrote: > > I started it off last time, so I'll take some of the blame... > > Past 2 weeks: > -adding unit tests to math.impl section > -getting reacquanted with linear algebra > > Next 2 weeks: > -finishing up math.impl tests > -adding tests to util.impl section > > -Hussein > > ----- Original Message ----- > From: "Bryan Tripp" <bp...@gm...> > To: "Development list for Java-based neural engineering simulator" > <nes...@li...> > Sent: Saturday, June 02, 2007 9:01 PM > Subject: Re: [Nesim-java] Friday, May 18 Update > > > > OK, everybody's off the hook for forgetting their updates on Friday, > > since I forgot too :) However, please send them in the next couple of > > days. > > > > Past 2 weeks: > > - cleanup of delay lines and impulse response decoding > > - finished plasticity framework > > - wrote spike-timing-dependent plasticity rule > > - experimented with multithreading in ensemble run > > > > Next 2 weeks: > > I may do some small tweaks in the next two weeks but I'll be working > > on something else for the most part. > > > > Bryan > > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by DB2 Express > > Download DB2 Express C - the FREE version of DB2 express and take > > control of your XML. No limits. Just data. Click to get it now. > > http://sourceforge.net/powerbar/db2/ > > _______________________________________________ > > Nesim-java mailing list > > Nes...@li... > > https://lists.sourceforge.net/lists/listinfo/nesim-java > > > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > Nesim-java mailing list > Nes...@li... > https://lists.sourceforge.net/lists/listinfo/nesim-java > -- Shu Wu http://www.sgsolutions.ca |
From: Hussein H. <hah...@uw...> - 2007-06-03 03:04:46
|
I started it off last time, so I'll take some of the blame... Past 2 weeks: -adding unit tests to math.impl section -getting reacquanted with linear algebra Next 2 weeks: -finishing up math.impl tests -adding tests to util.impl section -Hussein ----- Original Message ----- From: "Bryan Tripp" <bp...@gm...> To: "Development list for Java-based neural engineering simulator" <nes...@li...> Sent: Saturday, June 02, 2007 9:01 PM Subject: Re: [Nesim-java] Friday, May 18 Update > OK, everybody's off the hook for forgetting their updates on Friday, > since I forgot too :) However, please send them in the next couple of > days. > > Past 2 weeks: > - cleanup of delay lines and impulse response decoding > - finished plasticity framework > - wrote spike-timing-dependent plasticity rule > - experimented with multithreading in ensemble run > > Next 2 weeks: > I may do some small tweaks in the next two weeks but I'll be working > on something else for the most part. > > Bryan > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > Nesim-java mailing list > Nes...@li... > https://lists.sourceforge.net/lists/listinfo/nesim-java > |
From: Bryan T. <bp...@gm...> - 2007-06-03 01:01:58
|
OK, everybody's off the hook for forgetting their updates on Friday, since I forgot too :) However, please send them in the next couple of days. Past 2 weeks: - cleanup of delay lines and impulse response decoding - finished plasticity framework - wrote spike-timing-dependent plasticity rule - experimented with multithreading in ensemble run Next 2 weeks: I may do some small tweaks in the next two weeks but I'll be working on something else for the most part. Bryan |
From: Mike J. <jia...@gm...> - 2007-05-18 18:37:06
|
Hi all, Past week: -I have been doing a bit of reading about neuro engineering principles - getting java/junit set up while taking a look at the code. Next 2 weeks: - I will be looking into ca.neo.dynamics.impl classes to see what tests I can create. - I will also try to improve RK45intergrator to use previous initial time estimate. My goal aren't very specific/objective for this week. I think I will be able to give a better break down once I've done some actual coding. -Mike I will be looking into the On 5/18/07, Shu Wu <sh...@gm...> wrote: > > Past 2 weeks: > > - Getting used to new surroundings (in China) > - Set up with Nesim > - Progress made on NEOWorld (UI component of the simulator) > - Able to add NEO nodes to a surface from a canvas > > Next 2 weeks: > > - Allow users to create a IntegratorExample.java using GUI tools > > > -Shu > > > On 5/18/07, Bryan Tripp <bp...@gm...> wrote: > > > > Hi everyone, > > > > Past 2 weeks: > > - implemented transmission delays between neurons/ensembles (as a > > wrapper around Node) > > - using transmission delays to make a neural ensemble behave like a > > finite impulse response filter > > > > Nest 2 weeks: > > - clean up and commit above work > > - finish implementation of conversion from mixed-sign weight matrices > > to excitatory or inhibitory projection through interneurons (started > > in April) > > - start modelling spike-timing-dependent plasticity > > > > Bryan > > > > > > ------------------------------------------------------------------------- > > This SF.net email is sponsored by DB2 Express > > Download DB2 Express C - the FREE version of DB2 express and take > > control of your XML. No limits. Just data. Click to get it now. > > http://sourceforge.net/powerbar/db2/ > > _______________________________________________ > > Nesim-java mailing list > > Nes...@li... > > https://lists.sourceforge.net/lists/listinfo/nesim-java > > > > > > -- > Shu Wu > http://www.sgsolutions.ca > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > Nesim-java mailing list > Nes...@li... > https://lists.sourceforge.net/lists/listinfo/nesim-java > > -- Best Regards Mike Jiang Electrical Engineering Student University of Waterloo |
From: Shu W. <sh...@gm...> - 2007-05-18 14:23:38
|
Past 2 weeks: - Getting used to new surroundings (in China) - Set up with Nesim - Progress made on NEOWorld (UI component of the simulator) - Able to add NEO nodes to a surface from a canvas Next 2 weeks: - Allow users to create a IntegratorExample.java using GUI tools -Shu On 5/18/07, Bryan Tripp <bp...@gm...> wrote: > > Hi everyone, > > Past 2 weeks: > - implemented transmission delays between neurons/ensembles (as a > wrapper around Node) > - using transmission delays to make a neural ensemble behave like a > finite impulse response filter > > Nest 2 weeks: > - clean up and commit above work > - finish implementation of conversion from mixed-sign weight matrices > to excitatory or inhibitory projection through interneurons (started > in April) > - start modelling spike-timing-dependent plasticity > > Bryan > > ------------------------------------------------------------------------- > This SF.net email is sponsored by DB2 Express > Download DB2 Express C - the FREE version of DB2 express and take > control of your XML. No limits. Just data. Click to get it now. > http://sourceforge.net/powerbar/db2/ > _______________________________________________ > Nesim-java mailing list > Nes...@li... > https://lists.sourceforge.net/lists/listinfo/nesim-java > -- Shu Wu http://www.sgsolutions.ca |
From: Bryan T. <bp...@gm...> - 2007-05-18 14:12:55
|
Hi everyone, Past 2 weeks: - implemented transmission delays between neurons/ensembles (as a wrapper around Node) - using transmission delays to make a neural ensemble behave like a finite impulse response filter Nest 2 weeks: - clean up and commit above work - finish implementation of conversion from mixed-sign weight matrices to excitatory or inhibitory projection through interneurons (started in April) - start modelling spike-timing-dependent plasticity Bryan |
From: Hussein H. <hah...@uw...> - 2007-05-18 07:31:20
|
Hi everyone, I'm going to be working on adding unit tests to the math section. I had a little trouble with Maven not being able to find the jmatio and = jama jars when compiling, but it was simple enough to add them. I also changed the .classpath file to use a jar instead of a project for = JAMA. -Hussein |
From: Bryan T. <bp...@gm...> - 2007-05-15 17:04:33
|
Hi everyone, Welcome to our new mailing list. It may be a challenge to keep from stepping on each others' toes this summer with everyone so widely dispersed, but hopefully this will help. We'd like to try getting everyone to give regular updates on this list, so that we all have at least a rough sense of what the others are doing and can see potential conflicts or synergies. Let's do this every 2nd Friday starting this week. Just send a brief email listing what you have done on the simulator since last time, and what you plan to do in the following two weeks. I'll do this as well. I hope the list will also help me avoid answering questions multiple times. If you have a question about how the simulator works, please ask it here and I'll post the reply to everyone. Better yet, if you have an answer to someone else's question please jump in. Nobody is being evaluated on their questions, so don't hesitate to ask dumb ones. In fact, let's make a rule ... everyone must post at least two dumb questions by the end of the month. Joining us here is Marc, a PhD student who is learning the simulator. Marc, the guys we have working on development this summer are Shu, Mike, and Hussein. Bryan |