| --- Log opened Thu Jan 14 00:00:04 2016 | ||
| -!- sandeepkr [~sandeepkr@111.235.64.4] has joined ##hplusroadmap | 00:04 | |
| -!- jaboja64 [~jaboja@esa208.neoplus.adsl.tpnet.pl] has joined ##hplusroadmap | 00:12 | |
| -!- jtimon [~quassel@37.134.31.126] has joined ##hplusroadmap | 00:15 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has quit [Quit: Namaste] | 00:20 | |
| -!- |node [uid125132@gateway/web/irccloud.com/x-tzrpywwhuczmvwre] has quit [Quit: Connection closed for inactivity] | 00:21 | |
| -!- esmerelda is now known as mabel | 00:25 | |
| -!- wrldpc1 [~ben@hcccbcad3ab.bai.ne.jp] has joined ##hplusroadmap | 00:28 | |
| -!- drewbot [~cinch@ec2-54-158-121-155.compute-1.amazonaws.com] has quit [Remote host closed the connection] | 00:30 | |
| -!- drewbot [~cinch@ec2-54-82-72-168.compute-1.amazonaws.com] has joined ##hplusroadmap | 00:30 | |
| -!- Houshalter [~Houshalte@oh-71-50-56-224.dhcp.embarqhsd.net] has quit [Read error: Connection reset by peer] | 00:33 | |
| -!- jtimon [~quassel@37.134.31.126] has quit [Ping timeout: 272 seconds] | 00:44 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has joined ##hplusroadmap | 00:57 | |
| jrayhawk | kanzure: re: ikiwiki access: no; you could probably fake it imperfectly enough to throw indexing bots off your trail with some apache directives on the relevant paths. | 00:57 |
|---|---|---|
| jrayhawk | But cgit and ikiwiki.cgi can probably be made to expose stuff with by using POST to feed it variables. | 00:59 |
| -!- Filosofem [~Jawmare@unaffiliated/jawmare] has joined ##hplusroadmap | 01:02 | |
| -!- Jawmare [~Jawmare@unaffiliated/jawmare] has quit [Ping timeout: 260 seconds] | 01:06 | |
| -!- c0rw|zZz_ [~c0rw1n@101.154-67-87.adsl-dyn.isp.belgacom.be] has joined ##hplusroadmap | 01:15 | |
| -!- c0rw|zZz [~c0rw1n@101.154-67-87.adsl-dyn.isp.belgacom.be] has quit [Ping timeout: 276 seconds] | 01:17 | |
| -!- sandeepkr [~sandeepkr@111.235.64.4] has quit [Ping timeout: 260 seconds] | 01:45 | |
| -!- wrldpc1 [~ben@hcccbcad3ab.bai.ne.jp] has quit [Quit: wrldpc1] | 01:53 | |
| archels | "[...] for 52 years now, people thought CAD was an acronym for computer-aided design. But it is time to face facts: It really stands for computer-aided documentation." | 02:27 |
| archels | that's a cute take on it | 02:27 |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has quit [Remote host closed the connection] | 02:35 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has joined ##hplusroadmap | 02:35 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 02:40 | |
| -!- vicarion [~net@2001:388:608c:6cb5:d915:817:7729:d130] has quit [Ping timeout: 255 seconds] | 03:22 | |
| -!- pompolic [~A@unaffiliated/pompolic] has quit [Ping timeout: 255 seconds] | 03:25 | |
| -!- Gurkenglas [Gurkenglas@dslb-188-103-077-131.188.103.pools.vodafone-ip.de] has quit [Ping timeout: 264 seconds] | 03:36 | |
| -!- pompolic [~A@unaffiliated/pompolic] has joined ##hplusroadmap | 03:44 | |
| -!- jaboja64 [~jaboja@esa208.neoplus.adsl.tpnet.pl] has quit [Ping timeout: 264 seconds] | 03:49 | |
| -!- pompolic [~A@unaffiliated/pompolic] has quit [Ping timeout: 245 seconds] | 03:59 | |
| -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:9534:54a6:b815:c6f3] has quit [Read error: Connection reset by peer] | 04:00 | |
| -!- pompolic [~A@unaffiliated/pompolic] has joined ##hplusroadmap | 04:00 | |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has joined ##hplusroadmap | 04:09 | |
| chris_99 | this is awesome https://www.hodinkee.com/articles/a-major-step-forward-in-horological-additive-manufacturing-christoph-laimers-3d-printed-tourbillon | 04:25 |
| -!- jaboja64 [~jaboja@esa208.neoplus.adsl.tpnet.pl] has joined ##hplusroadmap | 04:30 | |
| -!- jaboja64 [~jaboja@esa208.neoplus.adsl.tpnet.pl] has quit [Ping timeout: 276 seconds] | 04:43 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 04:58 | |
| -!- berndj [~berndj@azna.co.za] has quit [Ping timeout: 240 seconds] | 05:02 | |
| -!- berndj [~berndj@azna.co.za] has joined ##hplusroadmap | 05:15 | |
| -!- berndj [~berndj@azna.co.za] has quit [Excess Flood] | 05:15 | |
| -!- berndj [~berndj@azna.co.za] has joined ##hplusroadmap | 05:16 | |
| -!- c0rw|zZz_ is now known as c0rw1n | 05:19 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has quit [Remote host closed the connection] | 05:38 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has joined ##hplusroadmap | 05:39 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 05:46 | |
| -!- Gurkenglas [~Gurkengla@dslb-188-103-077-131.188.103.pools.vodafone-ip.de] has joined ##hplusroadmap | 05:49 | |
| JayDugger | Good morning, everyone. | 06:01 |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has quit [Ping timeout: 265 seconds] | 06:01 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has quit [Remote host closed the connection] | 06:03 | |
| -!- atomical [~atomical@172.56.13.204] has joined ##hplusroadmap | 06:04 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has joined ##hplusroadmap | 06:07 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has quit [Ping timeout: 246 seconds] | 06:07 | |
| -!- |node [uid125132@gateway/web/irccloud.com/x-jonreqqhritlyawc] has joined ##hplusroadmap | 06:29 | |
| -!- wrldpc1 [~ben@p276100-ipngn200702kyoto.kyoto.ocn.ne.jp] has joined ##hplusroadmap | 06:42 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 06:47 | |
| -!- atomical [~atomical@172.56.13.204] has quit [Ping timeout: 255 seconds] | 06:49 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has quit [Excess Flood] | 06:57 | |
| -!- atomical [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 06:58 | |
| -!- atomical_ [~atomical@172.56.13.204] has joined ##hplusroadmap | 07:06 | |
| maaku | xentrac justanotheruser: everything that is wrong with AIXI | 07:09 |
| maaku | AIXI is basically "exhaustive/brute-force search for a solution". AIXItl is "make it technically computable by cutting off the brute-force search at some limit" | 07:10 |
| -!- atomical [~atomical@wl2142.cin.net] has quit [Ping timeout: 264 seconds] | 07:10 | |
| maaku | AIXI is of approximately zero interest to making an AGI because it is incomputable | 07:11 |
| maaku | AIXI-tl makes itself computable simply by setting a watchdog timer and terminating ... not interesting | 07:12 |
| maaku | it does nothing of interest to actually make sure that what is computed in the time available is of any relevance, and therefore would be hideously inefficient | 07:12 |
| maaku | and, no surprise, it's a publishable achievement when AIXI-tl is able to be used to solve even the simplest real world problems | 07:13 |
| maaku | xentrac: what are you trying to do that would have you look at AIXI? | 07:14 |
| poppingtonic | Even more inefficient would be doing inference on mcmc or particle cascade approximations of AIXItl's output. | 07:15 |
| poppingtonic | Also, AIXItl can't model itself as a part of the environment. | 07:16 |
| -!- sandeepkr [~sandeepkr@111.235.64.4] has joined ##hplusroadmap | 07:19 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-xliajdbzxtvnptmz] has quit [Quit: Connection closed for inactivity] | 07:20 | |
| poppingtonic | But it's not a waste of time to study it. | 07:31 |
| pasky | I also kinda like it as a theoretically AI-complete model that might be an inspiration for useful-something | 07:33 |
| pasky | as long as you're sure to understand that it's very far from applied research | 07:33 |
| -!- ArturShaik [~ArturShai@195.114.251.235] has quit [Ping timeout: 276 seconds] | 07:35 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has quit [Remote host closed the connection] | 07:46 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has joined ##hplusroadmap | 07:47 | |
| -!- ArturShaik [~ArturShai@212.241.17.179] has joined ##hplusroadmap | 07:48 | |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has joined ##hplusroadmap | 08:04 | |
| -!- poppingtonic [~Thunderbi@unaffiliated/poppingtonic] has quit [Ping timeout: 272 seconds] | 08:08 | |
| -!- atomical_ [~atomical@172.56.13.204] has quit [Ping timeout: 265 seconds] | 08:22 | |
| -!- atomical [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 08:24 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mwskgmkbnlwfijvj] has joined ##hplusroadmap | 08:34 | |
| -!- atomical [~atomical@wl2142.cin.net] has quit [Ping timeout: 255 seconds] | 08:41 | |
| -!- atomical [~atomical@172.56.13.204] has joined ##hplusroadmap | 08:42 | |
| -!- eudoxia [~eudoxia@r167-56-162-203.dialup.adsl.anteldata.net.uy] has joined ##hplusroadmap | 08:58 | |
| -!- balrog [~balrog@unaffiliated/balrog] has quit [Read error: Connection reset by peer] | 09:02 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 09:05 | |
| -!- c0rw1n [~c0rw1n@101.154-67-87.adsl-dyn.isp.belgacom.be] has quit [Remote host closed the connection] | 09:08 | |
| -!- atomical [~atomical@172.56.13.204] has quit [Ping timeout: 250 seconds] | 09:08 | |
| -!- c0rw1n [~c0rw1n@101.154-67-87.adsl-dyn.isp.belgacom.be] has joined ##hplusroadmap | 09:08 | |
| -!- eudoxia [~eudoxia@r167-56-162-203.dialup.adsl.anteldata.net.uy] has quit [Quit: Leaving] | 09:12 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 09:12 | |
| -!- balrog [~balrog@unaffiliated/balrog] has joined ##hplusroadmap | 09:15 | |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has quit [Ping timeout: 240 seconds] | 09:23 | |
| -!- jaboja [~jaboja@public-gprs384830.centertel.pl] has joined ##hplusroadmap | 09:23 | |
| -!- atomical [~atomical@172.56.13.237] has joined ##hplusroadmap | 09:26 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has quit [Ping timeout: 272 seconds] | 09:28 | |
| kanzure | CVE-2016-0777 - openssh leaks your private key to compromised boxes, time to upgrade | 09:36 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 09:55 | |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has joined ##hplusroadmap | 09:57 | |
| -!- sandeepkr [~sandeepkr@111.235.64.4] has quit [Ping timeout: 245 seconds] | 10:13 | |
| -!- jaboja [~jaboja@public-gprs384830.centertel.pl] has quit [Remote host closed the connection] | 10:16 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has joined ##hplusroadmap | 10:24 | |
| -!- nmz787_i [~ntmccork@134.134.139.76] has joined ##hplusroadmap | 10:37 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-mwskgmkbnlwfijvj] has quit [Quit: Connection closed for inactivity] | 10:39 | |
| kanzure | various vidoes https://www.tunnelsup.com/online-security-conferences/ | 10:42 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Remote host closed the connection] | 10:46 | |
| justanotheruser | maaku: I heard a rumor you're working on your own AGI system? | 10:54 |
| -!- c0rw1n is now known as c0rw|away | 11:02 | |
| maaku | justanotheruser: yes | 11:02 |
| maaku | for some definition of "working" | 11:02 |
| xentrac | maaku: basically pasky's take | 11:03 |
| maaku | justanotheruser: bitcoin/blockstream tends to take priority ... but that's really just to provide leverage to feed forward into an AGI-centered transhumanist project | 11:04 |
| -!- drethelin [~drethelin@24-241-226-112.dhcp.mdsn.wi.charter.com] has joined ##hplusroadmap | 11:04 | |
| justanotheruser | Is this the point where I suggest you make blockchain based AI? | 11:04 |
| * maaku shoots justanotheruser | 11:04 | |
| justanotheruser | lollll | 11:04 |
| justanotheruser | Is there a public repo, or papers, or is the project still in its infancy | 11:05 |
| maaku | in all seriousness people worried about AGI x-risk should be working on bitcoin now. we are -today- working on verifiable computing technology in adversarial environments while there is nothing but money on the line | 11:06 |
| justanotheruser | Yes, it would suck for an edge case to kill all humans | 11:07 |
| xentrac | maaku: that's a very good point | 11:08 |
| xentrac | maaku: how do you feel about the hardware and OS security landscape? | 11:08 |
| maaku | justanotheruser: Much better for an edge case to just result in theft and litigation! | 11:08 |
| xentrac | I finally installed a Ⓑ client | 11:08 |
| xentrac | well. I'm still 30 weeks behind on the blockchain | 11:09 |
| maaku | justanotheruser: also by happenstance the generalized script2.0 tech that people like gmaxwell and myself are interested in (it's my current research focus at blockstream) happens to overlap very nicely with the requirements for a base VM language in my AGI design | 11:11 |
| -!- |node [uid125132@gateway/web/irccloud.com/x-jonreqqhritlyawc] has quit [Quit: Connection closed for inactivity] | 11:11 | |
| maaku | xentrac: one of the things I'm involved with is open hardware and fully verified platforms | 11:12 |
| xentrac | so what's the state of the world? | 11:12 |
| xentrac | my impression is "we are up shit creek" but maybe there are things I'm not seeing | 11:13 |
| -!- Quashie_ [~boingredd@ool-18bccfe5.dyn.optonline.net] has joined ##hplusroadmap | 11:13 | |
| xentrac | not, any more, because we don't have the formal methods, but because we don't have access to manufacturing facilities | 11:14 |
| kanzure | if you are okay with large feature sizes then manufacturing is slightly more doable as a homebrew project | 11:15 |
| -!- Quashie [~boingredd@ool-18bccfe5.dyn.optonline.net] has quit [Ping timeout: 250 seconds] | 11:16 | |
| xentrac | I guess more generally I mean "because most people working on hardware are not working on open or verifiable hardware" | 11:17 |
| -!- ArturShaik [~ArturShai@212.241.17.179] has quit [Ping timeout: 265 seconds] | 11:19 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has quit [Read error: Connection reset by peer] | 11:21 | |
| maaku | RISC-V is looking like a good basis for full stack verification | 11:21 |
| xentrac | maaku: what does the possible paper path from verified Chisel to physical silicon look like? | 11:23 |
| maaku | Complicated :) | 11:23 |
| xentrac | does that mean "infeasible to make verifiable"? | 11:24 |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has joined ##hplusroadmap | 11:24 | |
| pasky | I'd expect it to be more fruitful for now to focus on making the thing actually generally intelligent in some rudimentary sense rather than work on vm languages etc. | 11:24 |
| maaku | Oh I agree I'm not an x-risker | 11:25 |
| maaku | But such VM languages is my day job and I've long since learned the value of double dipping work | 11:26 |
| xentrac | :D | 11:27 |
| Diablo-D3 | ssh client bug: http://undeadly.org/cgi?action=article&sid=20160114142733 | 11:30 |
| maaku | Such a VM is needed for my agi design regardless of x-risk | 11:34 |
| maaku | It is a good target for evolutionary search over program space guided by reasoning | 11:36 |
| xentrac | a verifiable-computing VM, maaku? | 11:37 |
| xentrac | Or what kind of a VM? | 11:37 |
| kanzure | a verified virtual machine that runs other verified software | 11:40 |
| -!- juul [~juul@juul.io] has quit [Quit: WeeChat 0.4.2] | 11:41 | |
| xentrac | well, so these are two different approaches | 11:41 |
| kanzure | unfortunately i would have to say i am an x-risker, i'm just the other side of it (the one that everyone else seems to be worried about) | 11:41 |
| xentrac | Bitcoin Script runs on a *verifiable* VM, in the sense that the computation gets replicated on other nodes, and they stochastically vote on whether to accept the results of the computation | 11:42 |
| nmz787_i | xentrac: open hardware is only closed if you're not on the inside | 11:42 |
| xentrac | in theory you could have multiple independent implementations of that VM so that a bug in one of the VMs would be caught by the others when they tried to replicate the computation | 11:42 |
| xentrac | although AFAIK nobody does | 11:42 |
| nmz787_i | I mean non-open | 11:43 |
| xentrac | but that's different from a *verified* VM where you've formally proven that the implementation fulfills its specification | 11:43 |
| xentrac | nmz787_i: ah, that makes much more sense :) | 11:43 |
| xentrac | I don't think it's actually true though because nobody is on all the insides | 11:43 |
| nmz787_i | I work at Intel | 11:44 |
| xentrac | You probably know a lot of things I don't, so I could be wrong | 11:44 |
| nmz787_i | but yeah, super big company, would take whole career to even get a decent overview as well as insight to innovate on ALL the processes | 11:44 |
| kanzure | why would you need to innovate on everything? | 11:45 |
| nmz787_i | hplus yo | 11:45 |
| maaku | Hah! I'm with kanzure. Destroy humanity! (and maybe replace it with something better) | 11:45 |
| nmz787_i | just name that AI humanity, and its still hplus | 11:46 |
| -!- juul [~juul@juul.io] has joined ##hplusroadmap | 11:46 | |
| maaku | Anyhow, my AGI design involves de novo concept learning using evolutionary search, guided by its knowledgebase. | 11:47 |
| nmz787_i | .wik silicaphobic | 11:47 |
| yoleaux | nmz787_i: Sorry, I couldn't find article. | 11:47 |
| maaku | "Guided by its knowledgebase" means probabalistic reasoning providing heuristics to guide search | 11:47 |
| maaku | "Verified software" means proving properties about code, which is a specific application of program reasoning. A VM designed to be easy to prove properties about is generally easy to reason about the behavior of programs, which makes it an easy base language for evolutionary search | 11:48 |
| kanzure | the other day you mentioned that you find humans too superstitious. but even with high levels of superstition, they seem to still do interesting computations that we have not yet figured out how to make computers do. doesn't this support the idea of reverse engineering some more? | 11:48 |
| kanzure | also, grounding problem stuff goes here | 11:48 |
| maaku | kanzure: This may be my own ignorance, but I sincerely don't know any neocortical algorithms that we can't do better or have ideas for doing better on a computer | 11:50 |
| kanzure | what is a neocortical algorithm? | 11:50 |
| maaku | I am interested in the old brain and expect to learn quite a bit from studies of that, including how our emotions work fundamentally | 11:50 |
| maaku | *emotions and drives | 11:50 |
| maaku | kanzure: things done by the neocortex | 11:50 |
| kanzure | neocortex has not been enough, so far, based on previous fidelity/resolution of simulations | 11:51 |
| maaku | kanzure: as you know my goal is to build an 'artificial scientist' / 'artificial engineer' . to THAT end I do not expect anything interesting to come out of brain science | 11:52 |
| kanzure | what level of evidence would be required to convince you that the brain reverse engineering strategy is better? | 11:54 |
| maaku | evidence that human beings are actually good at doing science :\ | 11:55 |
| kanzure | so far we are better than known computational implementations | 11:55 |
| kanzure | so..... | 11:55 |
| cluckj | we're the only ones doing it afaik | 11:55 |
| kanzure | indeed. | 11:55 |
| kanzure | well there's some partial implementations | 11:55 |
| kanzure | some weird stuff. like robot labs. but they have extremely limited scopes. | 11:55 |
| maaku | kanzure: not really. human beings augmented with non-human components do science -- we have methods and procedures external to ourselves that force good outcomes and censor bad behavior | 11:56 |
| kanzure | we do it better than a dog, and better than a markov bot. | 11:56 |
| maaku | i think we are talking past each other | 11:57 |
| cluckj | the machines we do science with don't have the same kind of agency that humans do | 11:57 |
| maaku | i just mean simple things like scientific rigor | 11:57 |
| kanzure | maaku: i think standards for convicing evidence is completely fair game. | 11:57 |
| cluckj | and we kinda....make those machines | 11:57 |
| maaku | that's not something that comes naturally to humans and is something we struggle hard with | 11:57 |
| maaku | cluckj: so we're fully general. that doesn't mean we're *good* at it | 11:57 |
| cluckj | I feel like "good" needs a point of comparison | 11:58 |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jxjfzzqgpamkfjip] has joined ##hplusroadmap | 11:59 | |
| maaku | cluckj: i can build a machine learning program to identify people by gait or voice. doesn't mean I'm good at it | 12:00 |
| kanzure | yes, so i think "evidence that humans are OK at it" is not a good response. | 12:00 |
| kanzure | identification is a falsehood- any data can mislead a program into identifying people by gait.... | 12:00 |
| maaku | kanzure: but if we do something well by building external processes ... then all i'm saying is cut humans out of the loop and we can move even faster | 12:01 |
| -!- Filosofem is now known as Jawmare | 12:10 | |
| -!- Gurkenglas [~Gurkengla@dslb-188-103-077-131.188.103.pools.vodafone-ip.de] has quit [Ping timeout: 255 seconds] | 12:17 | |
| -!- Quashie [~boingredd@ool-18bccfe5.dyn.optonline.net] has joined ##hplusroadmap | 12:22 | |
| -!- Quashie_ [~boingredd@ool-18bccfe5.dyn.optonline.net] has quit [Ping timeout: 255 seconds] | 12:25 | |
| xentrac | maaku: I agree with you about verified software, but what's the connection with script2.0? | 12:34 |
| xentrac | (maybe you've published notes somewhere on this that I should be reading instead of asking about it?) | 12:36 |
| xentrac | kanzure: is superstition the same as overfitting? | 12:36 |
| -!- vicarion [~net@2001:388:608c:6cb5:bcb1:7000:513c:e831] has joined ##hplusroadmap | 12:41 | |
| -!- Josh|NH4H [~quassel@botbot.xen.prgmr.com] has quit [Ping timeout: 240 seconds] | 12:42 | |
| -!- Josh|NH4H [~quassel@botbot.xen.prgmr.com] has joined ##hplusroadmap | 12:43 | |
| -!- nmz787_i [~ntmccork@134.134.139.76] has quit [Quit: Leaving.] | 12:51 | |
| pasky | maaku: I've been doing genetic programming and evolution a bit on and off and my lesson learned is that there's vast prior art and a lot to learn and *still* it's not useful for really high-dimensional problems and very tricky to get to work otherwise | 12:51 |
| pasky | so I hope you have some pretty clever ideas about how to make your evolutionary search work | 12:51 |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has joined ##hplusroadmap | 12:59 | |
| pasky | meanwhile, reading list while your models are training or evolving - http://phocks.org/stumble/creepy/ | 13:00 |
| xentrac | pasky: I was surprised to read in https://courses.cs.washington.edu/courses/csep573/11wi/lectures/04-lsearch.pdf yesterday that GAs don't in general perform better than hill climbing with random restarts | 13:01 |
| pasky | while that pdf is loading, what do you mean by "in general"? | 13:13 |
| maaku | xentrac: the need to verify the safety (software properties) of a contract before signing | 13:13 |
| maaku | e.g. can this script steal my money | 13:13 |
| maaku | pasky: nothing complex, just MOSES | 13:14 |
| maaku | pasky: the key is to not have an achitecture that relies on one mechanism for everything ;) | 13:14 |
| maaku | evolutionary program learning works for limited tasks. use other stuff for other tasks | 13:14 |
| pasky | so exactly what nontrivial tasks does evolutionary program learning work on? | 13:16 |
| pasky | xentrac: (for functions that aren't too highdimensional nor too lowdimensional, say 100-1000D, and have a very rugged landscape, methods based on genetic algorithms like BIPOP-CMA-ES are state-of-art; I think that's the most positive statement one can make about GA applicability) | 13:18 |
| -!- nmz787_i [ntmccork@nat/intel/x-jrmdxjosodvishiu] has joined ##hplusroadmap | 13:20 | |
| pasky | this is roughly the area where the popular fancy GA applications like antennas, mirrors, PCB routing and solar system trip planning fall | 13:21 |
| pasky | but it's fairly specific anyway | 13:21 |
| -!- nmz787_i1 [~ntmccork@134.134.139.76] has joined ##hplusroadmap | 13:23 | |
| -!- nmz787_i [ntmccork@nat/intel/x-jrmdxjosodvishiu] has quit [Ping timeout: 276 seconds] | 13:25 | |
| -!- maaku [~quassel@botbot.xen.prgmr.com] has quit [Quit: http://quassel-irc.org - Chat comfortably. Anywhere.] | 13:27 | |
| -!- maaku [~quassel@botbot.xen.prgmr.com] has joined ##hplusroadmap | 13:28 | |
| -!- maaku is now known as Guest82802 | 13:28 | |
| -!- Guest82802 is now known as maaku | 13:28 | |
| -!- vikraman [~vh4x0r@gentoo/developer/vikraman] has quit [Ping timeout: 260 seconds] | 13:29 | |
| -!- vikraman [~vh4x0r@gentoo/developer/vikraman] has joined ##hplusroadmap | 13:30 | |
| nmz787_i1 | has some interesting links, that are at least handy to be co-located on a single page:http://hackaday.com/2016/01/13/sourcing-your-cnc-tools-in-2016-build-them/ | 13:36 |
| FourFire | xentrac, making yourself bulletproof? | 13:56 |
| FourFire | > for implanting cloth under your chest skin, perhaps instead consider bio-compatible plastics | 13:56 |
| nmz787_i1 | there was a prior synbio art project that did something with transgenic spider silk | 13:58 |
| FourFire | ><Diablo-D3> and the problem with a lot of academic brute force optimization work seems the computer keeps finding solutions that do what you want by the rules you gave it, but are not useful solutions because it breaks other constraints you didn't know you had until it broke them | 13:58 |
| FourFire | semesl ike a good way to Learn. | 13:59 |
| kanzure | xentrac: superstition is... well maybe it's the same. see also: http://diyhpl.us/~bryan/papers2/ai/The%20evolution%20of%20superstitious%20and%20superstition-like%20behaviour.pdf | 14:01 |
| FourFire | > <xentrac> my impression is "we are up shit creek" but maybe there are things I'm not seeing | 14:04 |
| FourFire | My impresson as well | 14:04 |
| FourFire | I'm slightly exasperated that noone is picking the free money off the floor too ... | 14:05 |
| FourFire | or only a very few people are picking up only a bit of it | 14:05 |
| kanzure | xentrac: re: poor performance of GAs, consider using "lexicase selection" (selection based on multiple competing objective/utility functions) | 14:08 |
| kanzure | this is a topic that the beacon-center.org people wake up in the middle of the night about | 14:08 |
| kanzure | pasky: ref for failure of genetic algorithms to design antennas? i seem to recall matt campbell doing automatic antenna geometry design based on a* search algorithm. | 14:09 |
| maaku | up shit creek regarding what? | 14:14 |
| kanzure | luke-jr leaked recently that mindspillage is gmaxwell (and not kat walsh) http://gnusha.org/logs/mindspillage.txt | 14:19 |
| fenn | interesting.. i thought they had said more than that | 14:20 |
| kanzure | i'm guessing that luke-jr is wrong though :| | 14:21 |
| kanzure | https://pwnaccelerator.github.io/2016/sshbleed.html | 14:22 |
| maaku | pretty sure i've seen gmaxwell and mindspillage together in the same room :) | 14:23 |
| kanzure | yes but i haven't seen all *three* of you together in the same room | 14:23 |
| fenn | yes but (insert undecidable paraconsistent statement here) | 14:24 |
| kanzure | i can see that fenn is skilled at this | 14:24 |
| maaku | ah well you should have been at the office yesterday ;) | 14:24 |
| kanzure | you're aware that i don't work for you guys, right? | 14:24 |
| kanzure | i know it's hard to keep track :D | 14:24 |
| maaku | do'h! | 14:25 |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-rsrxbozmxferkaaz] has joined ##hplusroadmap | 14:41 | |
| fenn | from mindspillage.org: "Stealing a great statement from my partner Greg Maxwell (who is a cryptographer): “Information wants to be free” is the truth. " | 14:57 |
| kanzure | http://diyhpl.us/wiki/transcripts/gmaxwell-bitcoin-selection-cryptography/ | 15:00 |
| kanzure | " There was this rallying call, "information wants to be free". And I knew in my bones that this was true. We were going to use computers, which turn everything into information, and we would use networks to hook all of the computers together and we would change the world. We were going to change the power balances, make more people more empowered and everyone would have access to the world's knowledge and they would all fulfill their ... | 15:00 |
| kanzure | ... potential. That's a very political take on something that I now think in fact is better described as a law of nature. This is not just that I want information to be free; no, information really does want to be free. It is fundamental that information will percolate out into every little nook and cranny, and you can't control it. The result is that often bad things happen because information wants to be free. Sunlight is the ultimate ... | 15:00 |
| kanzure | ... solvent, but solvents corrode. So my email wants to be read by the NSA. When I try to login to my server, it can't tell me from you, because you can just replay my login. And now you're logged in as me. When you go and browse the internet, people learns how it works. They see inside your mind what used to be completely private. When I go to research something, marketers can send out cheap spam, and that spam is just as visible as the ... | 15:00 |
| kanzure | ... information I seek. If I want to build a digital cash system, I can't, because information is perfectly copyable. And all copies are just as good. Money that you can just copy is not much of a money at all. So you have an environment where there are powerful parties that have more ability to use this fundamental nature of information, and this goal of everyone being more empowered may not come true." | 15:00 |
| kanzure | btw, gmaxwell does not claim to be a cryptographer, in fact i think he always tends to drag a disclaimer nearby. | 15:01 |
| kanzure | oops maybe not. i don't have a quote from him on this. | 15:01 |
| fenn | * actually an ubercryptographer | 15:02 |
| kanzure | "i like to call myself more like a mathologist" | 15:02 |
| fenn | it's probably actually his backlash against "moon math" | 15:03 |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has quit [Quit: fencing] | 15:07 | |
| pasky | kanzure: i claimed the opposite; it's one of the tasks in suitable parameter range | 15:28 |
| pasky | no ref though, besides googling | 15:28 |
| maaku | that's a great quote from greg | 15:29 |
| -!- Filosofem [~Jawmare@unaffiliated/jawmare] has joined ##hplusroadmap | 16:06 | |
| -!- Jawmare [~Jawmare@unaffiliated/jawmare] has quit [Read error: Connection reset by peer] | 16:06 | |
| -!- nmz787_i1 [~ntmccork@134.134.139.76] has quit [Remote host closed the connection] | 16:25 | |
| -!- chris_99 [~chris_99@unaffiliated/chris-99/x-3062929] has quit [Quit: Leaving] | 16:54 | |
| -!- nmz787_i [~ntmccork@134.134.139.70] has joined ##hplusroadmap | 16:55 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 16:56 | |
| -!- atomical [~atomical@172.56.13.237] has quit [Ping timeout: 240 seconds] | 16:59 | |
| -!- Houshalter [~Houshalte@oh-71-50-56-224.dhcp.embarqhsd.net] has joined ##hplusroadmap | 16:59 | |
| -!- atomical [~atomical@172.56.13.237] has joined ##hplusroadmap | 17:04 | |
| -!- atomica__ [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 17:07 | |
| -!- atomical_ [~atomical@wl2142.cin.net] has quit [Ping timeout: 276 seconds] | 17:08 | |
| -!- atomical [~atomical@172.56.13.237] has quit [Ping timeout: 265 seconds] | 17:11 | |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-rsrxbozmxferkaaz] has quit [Quit: Connection closed for inactivity] | 17:20 | |
| xentrac | FourFire: right, for making myself somewhat bullet-resistant; bulletproof is too much to ask for | 17:22 |
| -!- night is now known as night|pub | 17:22 | |
| xentrac | kanzure: I am pretty sure that when I met mindspillage she was Kat Walsh and not gmaxwell, although they are not completely unrelated :) | 17:24 |
| -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:c5e2:fd8:3205:9ea3] has joined ##hplusroadmap | 17:31 | |
| nmz787_i | xentrac: you want it under the skin because??? you want to be able to tan while having security? | 17:39 |
| -!- nmz787_i [~ntmccork@134.134.139.70] has quit [Quit: Leaving.] | 17:53 | |
| -!- fleshtheworld [~fleshthew@2602:306:cf0f:4c20:c5e2:fd8:3205:9ea3] has quit [Disconnected by services] | 17:54 | |
| -!- fleshtheworld- [~fleshthew@2602:306:cf0f:4c20:c5e2:fd8:3205:9ea3] has joined ##hplusroadmap | 17:54 | |
| xentrac | nmz787: so I can't be coerced to remove it | 18:01 |
| -!- |node [uid125132@gateway/web/irccloud.com/x-yjvckhmwbrivoytw] has joined ##hplusroadmap | 18:04 | |
| kanzure | i'm sure a judge somewhere has demanded compulsory surgery | 18:08 |
| kanzure | oh, i bet that's happened in medical cases. so there you go. | 18:08 |
| -!- atomical [~atomical@172.56.13.240] has joined ##hplusroadmap | 18:19 | |
| -!- atomica__ [~atomical@wl2142.cin.net] has quit [Ping timeout: 264 seconds] | 18:23 | |
| xentrac | it happened frequently for eugenic reasons, as you should well know. | 18:26 |
| kanzure | indeed! i'm not sure your anti-surgery armor is going to work out favorably... | 18:27 |
| kanzure | oh right, you wanted surgery as the minimum barrier | 18:27 |
| xentrac | exactly | 18:27 |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 18:27 | |
| xentrac | no "don't you think it's a bit of an inappropriate statement to wear a bulletproof vest to this formal dinner?" or even "please, sir, remove your vest before entering the courtroom" or even "I've got a bad guy here in my scope, looks like he's wearing some kind of bulky vest, approaching the gate, permission to engage please?" | 18:29 |
| xentrac | coercion takes many forms | 18:31 |
| xentrac | very few of them are as expensive per person as surgery | 18:33 |
| AdrianG | compulsory surgery? | 18:57 |
| -!- mabel [~andares@unaffiliated/jacco] has quit [Ping timeout: 240 seconds] | 18:58 | |
| xentrac | no thanks | 18:59 |
| drethelin | http://www.israel-catalog.com/military-outdoor-gear/concealable-body-armor/bullet-proof-vest-sale/lightweight-bullet-proof-waistcoat-level-3-a | 19:01 |
| drethelin | they make bulletproof vests that look formal nowadays | 19:01 |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Read error: Connection reset by peer] | 19:04 | |
| -!- atomical_ [~atomical@172.56.13.240] has joined ##hplusroadmap | 19:04 | |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 19:05 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has joined ##hplusroadmap | 19:06 | |
| -!- atomica__ [~atomical@wl2142.cin.net] has joined ##hplusroadmap | 19:07 | |
| -!- atomical [~atomical@172.56.13.240] has quit [Ping timeout: 264 seconds] | 19:08 | |
| -!- esmerelda [~andares@unaffiliated/jacco] has joined ##hplusroadmap | 19:09 | |
| -!- atomical_ [~atomical@172.56.13.240] has quit [Ping timeout: 272 seconds] | 19:11 | |
| -!- jtimon [~quassel@126.31.134.37.dynamic.jazztel.es] has quit [Ping timeout: 246 seconds] | 19:30 | |
| -!- ArturShaik [~ArturShai@212.241.17.179] has joined ##hplusroadmap | 19:33 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has quit [Read error: Connection reset by peer] | 19:44 | |
| -!- yash [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has joined ##hplusroadmap | 19:45 | |
| -!- erasmus [~erasmus@unaffiliated/erasmus] has joined ##hplusroadmap | 19:46 | |
| -!- yashgaroth [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Ping timeout: 250 seconds] | 19:48 | |
| -!- dcentral [~IGLC@2601:680:c400:e360:938:5f5b:6421:e8ef] has joined ##hplusroadmap | 20:13 | |
| -!- dcentral [~IGLC@2601:680:c400:e360:938:5f5b:6421:e8ef] has quit [Client Quit] | 20:13 | |
| -!- pompolic [~A@unaffiliated/pompolic] has quit [Read error: Connection reset by peer] | 20:14 | |
| -!- pompolic [~A@unaffiliated/pompolic] has joined ##hplusroadmap | 20:14 | |
| -!- Burnin8 is now known as Burninate | 20:54 | |
| -!- sandeepkr [~sandeepkr@111.235.64.4] has joined ##hplusroadmap | 21:11 | |
| -!- nmz787_i [ntmccork@nat/intel/x-rueakcatoqgcgdrz] has joined ##hplusroadmap | 21:47 | |
| -!- yash [~yashgarot@2602:306:35fa:d500:f5e0:f867:a11d:8d52] has quit [Quit: Leaving] | 21:48 | |
| Diablo-D3 | http://blog.robertelder.org/virtual-memory-with-256-bytes-of-ram/ | 21:59 |
| -!- PatrickRobotham [uid18270@gateway/web/irccloud.com/x-iwiwfcxmwysznswu] has joined ##hplusroadmap | 22:24 | |
| -!- jackybgood [~jackybgoo@98-125-92-211.dyn.centurytel.net] has joined ##hplusroadmap | 22:32 | |
| -!- Jaw_ [~Jawmare@unaffiliated/jawmare] has joined ##hplusroadmap | 22:35 | |
| -!- Filosofem [~Jawmare@unaffiliated/jawmare] has quit [Ping timeout: 256 seconds] | 22:39 | |
| -!- jackybgood [~jackybgoo@98-125-92-211.dyn.centurytel.net] has quit [Remote host closed the connection] | 22:47 | |
| -!- night|pub is now known as night | 23:18 | |
| -!- |node [uid125132@gateway/web/irccloud.com/x-yjvckhmwbrivoytw] has quit [Quit: Connection closed for inactivity] | 23:41 | |
| -!- CheckDavid [uid14990@gateway/web/irccloud.com/x-jxjfzzqgpamkfjip] has quit [Quit: Connection closed for inactivity] | 23:49 | |
| -!- Houshalter [~Houshalte@oh-71-50-56-224.dhcp.embarqhsd.net] has quit [Read error: Connection reset by peer] | 23:56 | |
| --- Log closed Fri Jan 15 00:00:05 2016 | ||
Generated by irclog2html.py 2.15.0.dev0 by Marius Gedminas - find it at mg.pov.lt!