6 ms·

Machine Learning Unconference

125 points by albertzeyer 10y ago

qwertyuiop924 10y ago
Bryan Cantrill's USENIX talk "The Emperor's Wardrobe" discussed a lot of the issues that the Unconference is trying to solve.
- ihaveajob 10y ago
  Do you have a link to the talk, or a reference to it? It seems interesting.
  - qwertyuiop924 10y ago
    Certainly, good sir. ;-) https://www.usenix.org/conference/atc16/technical-sessions/presentation/cantrill https://www.usenix.org/conference/atc16/technical-sessions/p... There's audio and slides. No video, unfortunately.
biofox 10y ago
Sounds like this might be interesting, but I have no idea why unconferences are growing in popularity. In my experience the only differences to conferences is that they are poorly organised and have fewer high calibre attendees. Am I missing something?
- msutherl 10y ago
  Makes more sense when it's for and by an actual community of people, in which case it's a bit more like a party with your friends where you all give talks and take part in semi-structured activities
ThePhysicist 10y ago
Sounds interesting! I have been trying to get in touch with the OpenAI team in the last weeks concerning their program synthesis project (I have been working on static analysis & machine learning for the last two years and would be very eager to contribute some ideas) but so far haven't been successful. It seems they're either just unresponsive or overwhelmed by requests right now (I assume the latter). Meeting the team at the conference seems like a great opportunity but I'm wondering whether this will be an effective way to get a collaboration started or learn how to participate in their efforts (as it would require a 20 hour trip from Berlin for me). For me, the biggest questions right now are: * How are they going to organize their work? * What kind of contributions are they looking for? I think it would be great to address these things at the conference.
- yazr 10y ago
  Just had a look at your presentation "Learning from other's mistakes: Data-driven code analysis" Can you share some links to your work with ML or related ?
  - ThePhysicist 10y ago
    We have a solution for static analysis of Python code online: https://www.quantifiedcode.com https://www.quantifiedcode.com Currently we're using machine learning mostly behind the scenes to optimize our hand-written patterns for analysis and refactoring, but we started working on a deep learning based system to improve our analysis, using data we've collected from the 10.000 software projects we are analyzing right now.
    - bearzoo 10y ago
      I would email greg brockman. I pestered him for advice one time and the amount of time he took to respond (and the amount of times he responded) was well beyond my expectation.
- ilya_su 10y ago
  Ilya from OpenAI here. We are swamped indeed. But please ping me at ilyasu@openai.com and I will get back to you.
fitzwatermellow 10y ago
Interesting discussion on the Gitter about livestreaming the unconference vs online participation vs freenode #ml channels vs tensorflow slack rooms vs /r/MachineLearning. But actually, it seems the Gitter for OpenAI itself is a great place to start! Maybe someone could volunteer to provide "live notes" of the Unconference on this channel?
mindcrime 10y ago
This sounds really exciting. SFO is a long haul for me (coming from RDU), but I applied and am hoping to get in. I'll definitely attend if possible. Hopefully they make decisions pretty quick, so people can book flights while they're still cheap(ish).
- Chronic9q 10y ago
  Did you look at the list of accepted attendees? Unless you can send a SMS text to someone at OpenAI, you're not getting in.
  - mindcrime 10y ago
    Probably, but all one can do is apply and hope for the best.
20yrs_no_equity 10y ago
I like the idea of a timely conference, but it feels like the whole idea of conferences is a bit dated. Sure unconference means that it is more informal and in this case more timely, but I'm not flying across the country for a few days. Shouldn't we, as the general set of hackers, be developing or exploring online virtual conferences? A Speaker in Norway in front of a camera presenting their stuff to thousands of people around the world. Has this been tried? Is there a good platform for it? This seems much more efficient- you can have conferences more easily and frequently without the interruption and cost of having to physically be there. In my team there are a dozen global tech conferences that are relevant to us, but only the sales guys and "executives" at my company go to them. The engineers do not. Because it costs us thousands of dollars to send someone to shanghai. And it's perceived that the sales guy makes sense to send, but not the engineer, cause we don't need to spend thousands on keeping engineers up to date with technology, or have them make cross relationships with others advancing the state of the art.
- gary_bernhardt 10y ago
  There are online conferences, but they're very different from the in-person ones. Here are some properties of in-person conferences that are difficult to replace. * Some talks are long, intricate stories, and you get much more out of them by paying close attention. (David Beazley comes to mind.) From experience, I can say that I'm far more likely to divide my attention at home than in an in-person talk, and I'm not unusual here. * Sometimes, there's no talk that grabs you in a time slot, but that can lead to pleasant surprises. This is especially effective if you intentionally choose topics that are unfamiliar, then let the physicality of the room force you to pay attention (because it would be rude not to). * Running into people randomly in the hall makes for serendipitous conversations that wouldn't have happened otherwise. People get jobs this way; open source projects start this way. The global Code Retreat movement started this way due to a conversation at CodeMash years ago. (That conversation began by us talking about the ethics of the meat in the sandwiches.) * If there are people who you put on a pedestal as untouchable divine entities, meeting them will usually fix that. I don't mean that they don't deserve respect, but simply that everyone is human. You might be appalled to learn that your favorite programming tool's author likes music that you find repugnant. This is valuable; we're people. You'll be less likely to appeal to authority unthinkingly if you've been appalled by authorities' taste in music. * It's vulgar to put this so bluntly, but a lot of power is brokered in person at conferences. Not through mysterious back channel conspiracies, but through normal human interactions. If I've met you in person, I'm more likely to trust you; you're more likely to come to mind when someone asks "do you know anyone who can do x?"; etc. Anyone who doesn't participate is at a disadvantage here (and many surely don't care, which is fine). (Source: I've spoken at about 40 conferences, give or take, and attended many without speaking as well; I've made a lot of friends who do the same; I shot an unpublished documentary that was sort of about this; and I have repugnant taste in music.)

125 points by albertzeyer 10y ago

Bryan Cantrill's USENIX talk "The Emperor's Wardrobe" discussed a lot of the issues that the Unconference is trying to solve.

ihaveajob 10y ago
Do you have a link to the talk, or a reference to it? It seems interesting.
- qwertyuiop924 10y ago
  Certainly, good sir. ;-) https://www.usenix.org/conference/atc16/technical-sessions/presentation/cantrill https://www.usenix.org/conference/atc16/technical-sessions/p... There's audio and slides. No video, unfortunately.

biofox 10y ago

Sounds like this might be interesting, but I have no idea why unconferences are growing in popularity. In my experience the only differences to conferences is that they are poorly organised and have fewer high calibre attendees. Am I missing something?

msutherl 10y ago
Makes more sense when it's for and by an actual community of people, in which case it's a bit more like a party with your friends where you all give talks and take part in semi-structured activities

ThePhysicist 10y ago

Sounds interesting! I have been trying to get in touch with the OpenAI team in the last weeks concerning their program synthesis project (I have been working on static analysis & machine learning for the last two years and would be very eager to contribute some ideas) but so far haven't been successful. It seems they're either just unresponsive or overwhelmed by requests right now (I assume the latter). Meeting the team at the conference seems like a great opportunity but I'm wondering whether this will be an effective way to get a collaboration started or learn how to participate in their efforts (as it would require a 20 hour trip from Berlin for me). For me, the biggest questions right now are: * How are they going to organize their work? * What kind of contributions are they looking for? I think it would be great to address these things at the conference.

yazr 10y ago
Just had a look at your presentation "Learning from other's mistakes: Data-driven code analysis" Can you share some links to your work with ML or related ?
- ThePhysicist 10y ago
  We have a solution for static analysis of Python code online: https://www.quantifiedcode.com https://www.quantifiedcode.com Currently we're using machine learning mostly behind the scenes to optimize our hand-written patterns for analysis and refactoring, but we started working on a deep learning based system to improve our analysis, using data we've collected from the 10.000 software projects we are analyzing right now.
  - bearzoo 10y ago
    I would email greg brockman. I pestered him for advice one time and the amount of time he took to respond (and the amount of times he responded) was well beyond my expectation.
ilya_su 10y ago
Ilya from OpenAI here. We are swamped indeed. But please ping me at ilyasu@openai.com and I will get back to you.

fitzwatermellow 10y ago

Interesting discussion on the Gitter about livestreaming the unconference vs online participation vs freenode #ml channels vs tensorflow slack rooms vs /r/MachineLearning. But actually, it seems the Gitter for OpenAI itself is a great place to start! Maybe someone could volunteer to provide "live notes" of the Unconference on this channel?

mindcrime 10y ago

This sounds really exciting. SFO is a long haul for me (coming from RDU), but I applied and am hoping to get in. I'll definitely attend if possible. Hopefully they make decisions pretty quick, so people can book flights while they're still cheap(ish).

Chronic9q 10y ago
Did you look at the list of accepted attendees? Unless you can send a SMS text to someone at OpenAI, you're not getting in.
- mindcrime 10y ago
  Probably, but all one can do is apply and hope for the best.

20yrs_no_equity 10y ago

I like the idea of a timely conference, but it feels like the whole idea of conferences is a bit dated. Sure unconference means that it is more informal and in this case more timely, but I'm not flying across the country for a few days. Shouldn't we, as the general set of hackers, be developing or exploring online virtual conferences? A Speaker in Norway in front of a camera presenting their stuff to thousands of people around the world. Has this been tried? Is there a good platform for it? This seems much more efficient- you can have conferences more easily and frequently without the interruption and cost of having to physically be there. In my team there are a dozen global tech conferences that are relevant to us, but only the sales guys and "executives" at my company go to them. The engineers do not. Because it costs us thousands of dollars to send someone to shanghai. And it's perceived that the sales guy makes sense to send, but not the engineer, cause we don't need to spend thousands on keeping engineers up to date with technology, or have them make cross relationships with others advancing the state of the art.

gary_bernhardt 10y ago
There are online conferences, but they're very different from the in-person ones. Here are some properties of in-person conferences that are difficult to replace. * Some talks are long, intricate stories, and you get much more out of them by paying close attention. (David Beazley comes to mind.) From experience, I can say that I'm far more likely to divide my attention at home than in an in-person talk, and I'm not unusual here. * Sometimes, there's no talk that grabs you in a time slot, but that can lead to pleasant surprises. This is especially effective if you intentionally choose topics that are unfamiliar, then let the physicality of the room force you to pay attention (because it would be rude not to). * Running into people randomly in the hall makes for serendipitous conversations that wouldn't have happened otherwise. People get jobs this way; open source projects start this way. The global Code Retreat movement started this way due to a conversation at CodeMash years ago. (That conversation began by us talking about the ethics of the meat in the sandwiches.) * If there are people who you put on a pedestal as untouchable divine entities, meeting them will usually fix that. I don't mean that they don't deserve respect, but simply that everyone is human. You might be appalled to learn that your favorite programming tool's author likes music that you find repugnant. This is valuable; we're people. You'll be less likely to appeal to authority unthinkingly if you've been appalled by authorities' taste in music. * It's vulgar to put this so bluntly, but a lot of power is brokered in person at conferences. Not through mysterious back channel conspiracies, but through normal human interactions. If I've met you in person, I'm more likely to trust you; you're more likely to come to mind when someone asks "do you know anyone who can do x?"; etc. Anyone who doesn't participate is at a disadvantage here (and many surely don't care, which is fine). (Source: I've spoken at about 40 conferences, give or take, and attended many without speaking as well; I've made a lot of friends who do the same; I shot an unpublished documentary that was sort of about this; and I have repugnant taste in music.)