[ / / / / / / / / / / / / / ] [ dir / agatha2 / animu / doomer / eris / s8s / tingles / vg / vichan ]

/bant/ - Le Dank Forced Face/Random

ℂ⌬©⍟ℕυͳ's shitposting ground
Winner of the 80rd Attention-Hungry Games
/otter/ - Otter For Your Soul

Entries for the 2019 Summer Infinity Cup are now open!
May 2019 - 8chan Transparency Report
Name
Email
Subject
Comment *
File
Password (Randomized for file and post deletion; you may also set your own.)
* = required field[▶ Show post options & limits]
Confused? See the FAQ.
Flag
Embed
(replaces files and can be used instead)
Oekaki
Show oekaki applet
(replaces files and can be used instead)
Options
dicesidesmodifier

Allowed file types:jpg, jpeg, gif, png, webm, mp4, swf, pdf
Max filesize is 16 MB.
Max image dimensions are 15000 x 15000.
You may upload 5 per post.


Consider checking out :- [8/bants/] [⑨/nap/]

File: 446f3b6fcbe1fce⋯.png (637.56 KB, 3840x2500, 192:125, comfy mode.png)

e2ccae  No.481

I will respond to everyone in this thread.

3a4ca5  No.482

ass flaps


3a4ca5  No.483

File: ab95b64b5325cde⋯.jpg (106.05 KB, 790x790, 1:1, Zv8sX1i.jpg)


a93966  No.484

>>482

Ass flaps are what I imagine is the correct way to describe an elderly person butt

>>483

*Claps


8c3b2d  No.485

What is your favourite colour?

Which Homestuck would you suck?


a93966  No.486

>>485

My favorite color is green like my eyes. I'm not sure what Homestuck is


3f5db9  No.487

File: b0845a8c7913ae5⋯.jpg (52.05 KB, 500x674, 250:337, tumblr_lpiwqw3WyJ1r18quno1….jpg)

>>486

>I'm not sure what Homestuck is

All things considered, it's probably best to keep it that way.


8bbf84  No.490

How is your life going


a93966  No.491

>>490

Fug I let this post go a whole day without responding

Pretty much unchanged. I think I want to go back to school, and get a bachelor's in electronics. Just turned 26 the other day, On the fast track to wizardom. I think I got depression under control, just have to cycle 24 miles a day. This city sucks for riding a bike. Rednecks yell at me everyday. Today one slowed down to yell something at me, I just smiled and and jerked the bike sideways like I was going to run into his car. He nearly had an accident. Pretty funny. One of my d

Freinds from NC tracked me down. If you don't remember I just left without talking to anyone. We had a good chat and I honestly enjoyed it, but then came that closterphobic feeling I get when people are to familiar. There is something wrong with me I don't know why this happens. I just want to run.

Also I was going to draw a pic for every reply ,but even that shitty OP pic took me like a whole day. That shits hard, and I'm not the artist type.


3a4ca5  No.492

>>491

lole, it's disturbing how familiar this story sounds. almost word for word, except for the rednecks

I got away from everyone except this one guy that insists on seeing me like once a year

Keeps making me remember the person I used to be

School is worth considering, I did

Good luck


5bb06a  No.493

>>492

It's like 50/50 go to school or become a hobo again.


3a4ca5  No.495

>>493

Easy choice to make then. If you've had a hard time working for a long time getting some degree is like the easiest way of having some proof to say "look I'm committed". Altho obviously school is more fun if you have something more on your mind than just work but still


5bb06a  No.496

>>495

I already have a 2 year degree and more importantly I got a good job in the field. I feel like I've already proven that. I loved what I do and study in my free time. School is just soul crushing. I'm the type that can only learn if I teach myself, so school seems even more pointless.


3a4ca5  No.497

>>496

So why did you say you wanna go back in the first place? That mindset is a tad deceptive tho. Many people believe themselves great autodidacts. I genuinely thought that too until I actually went back to uni and had to realize the hard way my homebrew methods of learning were horribly inefficient and the information I could find by myself was sorely lacking in quality. But you're not me and I am an idiot, it's entirely possible what you're saying is true. But just make sure you're confident of your own learning skills if you stick to it or you could end up missing out


5bb06a  No.498

>>497

I need that piece of paper so I can get that paper. As for resources, Stanford university posts all thier classes on YouTube. I've been working my way through some of their AI courses. I'm pretty thorough I've spent 2 weeks making sure I understood everything in the first lecture. I had to work my way through an entire vector calculus and statistics course just to get caught up.


3a4ca5  No.499

File: 9cbba1834ec5cc1⋯.png (1.45 MB, 1050x903, 50:43, 9cb.png)

>>498

Neat. AI is kewl, I'm thinking of taking it as a bonus degree. There's so many topics in there what exactly did you do? I remember talking about it with the local AI professor. I asked him about statistics specifically because I kinda hate it as opposed to anything else, and funny enough he mentioned they don't spend much of their research actually doing much statistics. At any rate if you ever need anything mathrelated hmu. I'm very bad at everything in life but math seems to be the one exception. I have quite a few books at my disposal but obviously most of them aren't in english


48567c  No.502

>>499

I want to learn about neural networks and natural language processing. I want to make a website the scrapes news websites, and attempt to remove bias and constuct the most consistent series of events.

It's an intersection of mathmatics and computing, so the choice of algorithm you use is based mostly on computing efficiency. Which means it comes down to experimentation bits of mathmatics until it just werks. I have to do alot of catching up before I could understand it fully. It's one of those fields where it seems like people are really still poking around in the dark and anyone could stumble across something important.


3a4ca5  No.506

>>502

Heh. I've been pretty mad at how powerless we are to the current shitstorm of skewed and unnecessary news going around too.

The rules of detecting and fixing bias are vague, ambiguous and sometimes biased themselves. Creating an automated system that can't just read, but actually interpret text, *and* derive correct action and consequence from multiple, sometimes contradictory sources is from what I can tell pretty much uncharted territory. Computational efficiency is gonna be the least of your worries until you've actually got an idea how the system is supposed to work


48567c  No.511

>>506

There has been a bit of research on into the subject. Turns out neural networks are really good at detecting negative or positive sentiments. So why not bias? Another way news organizations show bias is based on what they choose to cover. Machine learning is also good at discovering the subject of a body of text. I could then impliment a set of rules to say what's important and what's not. Also there is the problem of removing my own bias from the system.


3a4ca5  No.513

>>511

Not too sure man. I'm just kinda thinking bias is quite a bit more subtle than sentiment. Sentiment is a direct response of text, completely independent of any other text involved, while bias not only requires knowledge of the text itself, but of whatever its content is dealing with, requires cross referencing with texts on the same topic both to assert and to correct, can acquire a cumulative effect from multiple sources continuously playing on eachother's talking points, etcetera. I highly doubt it's even remotely as easy relative to sentiment. From what I can tell having no personal bias is composed of three basic principles. First you gotta be able to read words at a very strict face value and nothing more. Every word in a sentence matters. Lots of times when people lose their shit over something it seems they glossed over very specific words in a sentence that make a lot of difference. Secondly you need a strong sense of deductive logic, making sure to only derive information that can be asserted with absolute certainty. And last you need to be able to represent this information in words again, making sure it is both complete and unambiguous. I'm pulling all this out of my ass obviously but that's about what I make of it


9bd730  No.515

>>513

That's why it's important to be able to pick out events. X person went here. He did this. Then he said X. It's kinda hard to be bias with that sort of thing. I've seen a demo by some researches that did just that. It created a tree sort of graph out of Romeo and Juliet. Containing just the bare cliff notes version of what happened and how it was connected. Also Keep in mind non bias doesn't mean correct I'm not trying to make a truly self aware all knowing AI here. It will just gather the daily news compare stories from multiple sources and give a very dry summary. Leaving out anything that sounds like an opinion.


3a4ca5  No.516

>>515

Oh yeah that makes more sense than whatever I had in mind. Still won't be easy but

best of luck to you at any rate. I really dig this kind of stuff so if you ever stumble into anything interesting keep us posted


8dcdad  No.593

Hello


8bbf84  No.594

Response get




[Return][Go to top][Catalog][Nerve Center][Cancer][Post a Reply]
Delete Post [ ]
[]
[ / / / / / / / / / / / / / ] [ dir / agatha2 / animu / doomer / eris / s8s / tingles / vg / vichan ]