Closed hdmchl closed 8 years ago
also... how do I label this issue with "Melbourne"?
I'm interested
We could definitely do something similar in SYD - I should see if I can find a "node-sympathetic" ML researcher from the NICTA machine learning research team.
hi All,
On 2014-06-22 21:44 , Hadi Michael wrote:
Is anyone interested in getting involved in a machine learning / AI focused ...
I'm very interested, particularly regarding computer vision.
However personally, for the next NBD ... will still be concentrating on getting the hardware, e.g SimpleBot, networking and underlying software to be better / fastest for the attendees.
Very likely to have a small camera on-board the next SimpleBot iteration.
-O- cheers = /\ /\/ /) `/ = --O -- http://www.geekscape.org -- OOO -- andyg@geekscape.org -- http://twitter.com/geekscape --
Cool, since we have interest, the next step is to brainstorm some ideas. What would we build / do on the day? Here are a couple of ideas to get things going:
@geekscape: @sidorares and I can facilitate the session. Would be good to get your thoughts on all this though... @DamonOehlman: is there anyone at NICTA that you can get me in touch with? Would be good to see what other people have been working on - or more specifically, what they're using ML to solve.
Hey peeps.
So I think if we're going to cover ML then it should be applicable to the NodeBots side of thing. Product recommendations and such like is a good basic intro to ML as it introduces bayesian ideas which is a good foundation point however I'd prefer if we keep it relevant for peeps participating on the day.
(Aside FYI the DataScience meetup is excellent for this sort of stuff)
So on that basis I'd consider things like:
Don't forget that as we go down this path we need the bots to be more autonomous so need to consider the autonomy component of this sort of thing. I'm happy for peeps to play with my roverbot thing which has full RPI / Arduino / Battery type set up to explore this sort of stuff and would give me a good excuse to get it working again as I've been so SimpleBot focussed of late.
Also note that the vast majority of peeps coming along will be newbies to robotics etc so this might be something that is worth kicking out for ongoing stuff as part of the ENR sessions at HackMelb. But an intro to the subject matter never hurts!
Anyway, that's just my view because as much as I'd love to be doing autonomous vehicles and things like that I suspect most of the participants will have very modest goals (make an LED blink, drive a bot, get a range finder working)... as we've seen at many of the events to date.
Andrew Fisher http://about.me/ajfisher/bio
On Mon, Jun 23, 2014 at 9:10 PM, Hadi Michael notifications@github.com wrote:
Cool, since we have interest, the next step is to brainstorm some ideas. What would we build / do on the day? Here are a couple of ideas to get things going:
- Self-driving simplebot?
- Product recommendations, ML using node.js?
@geekscape https://github.com/geekscape: @sidorares https://github.com/sidorares and I can facilitate the session. Would be good to get your thoughts on all this though... @DamonOehlman https://github.com/DamonOehlman: is there anyone at NICTA that you can get me in touch with? Would be good to see what other people have been working on - or more specifically, what they're using ML to solve.
— Reply to this email directly or view it on GitHub https://github.com/nodebotsau/nbdau/issues/26#issuecomment-46831028.
Looks like we may finally make a start on this ... this year !
Hi all, is anyone interested in getting involved in a machine learning / AI focused breakout session at NBD? Do people think this is something that would be appropriate for the day?
I'm keen to see a stronger focus on the software side of robotics. I'm thinking we can use ML to learn something from physical sensors and then have the bot react accordingly. Nothing too crazy. The core objective would be to get people thinking about connecting the software to the hardware.
I'm looking at libraries like: Neural network learning: https://github.com/harthur/brain Natural language detection: https://github.com/NaturalNode/natural ++ all the computer vision stuff that we could integrate...