#machine learnign

[alexparozi]

sooo happy!

[ask-mithrun]

have you ever tried sewing/knitting/crochet?

No, i haven't.

Perhaps i can try for the sake of "new desires" but its a little too far from my prexisting hobbies so the learning curve might be seep. Some day maybe.

I never really had the reason to make my own clothing or things adjacent to it any ways, I Usually commission from professional seemstresses for my garments — i have the money for that.

[karatekamania]

as someone going into science, i like to imagine what the average day in love lab is like

[txttletale]

whenever i see people talk about 'joke theft' in the comedy world as a comparison for AI art while at the same time pushing for copyright regulation -- i mean, for one, it's a completely incoherent comparison, there's just no similarity whatsoever to how machine learnign models actually work -- but even if it was, do you have any fucking idea how much worse standup comedy would be if 'joke theft' was a legal concept enshrined in copyright law and not just something considered a bit of a dick move?

like can you imagine a world where jerry seinfeld or ricky gervais could use their millions of dollars to sue anyone who makes a joke that happens to bear resemblance to one of theirs (because they're riffing on a similar topic or making a similar common observation), while being able to easily drag out or even settle out of court any lawsuits filed against them with those same millions? a world where the standup aristocracy are even more entrenched than they already are? it's like people don't even realize the meaning of the sentences they say about this shit

[valdangelodreams]

AU #12 (Nona Di Angelo)

I could not tell you where this one came from, but then, most of these are me getting dragged by my unconscious into funky stories

-anyway, mortal au

-Nico's family is him, Hazel, and their grandma

-nona di angelo is only really related to Nico, but she loves Hazel

-obvs the sibs love their nona, she's the best

-however, with Hazel on college and Nico working several jobs to support nona, neither sibling sees her much

-Nico reaches out to Reyna, who reaches out to Piper who takes care of her in small ways (she can handle herself, but she's an eldery woman and her body doesn't always agree with her)

-one day though, the lights go out on them, and if they call for professional help they might not get fixed any time soon

-so Piper calls Leo, who's there in seconds after her short explanation, and then he has the lights fixed in minutes

-he's invited to stay for dinner, and only agrees after nona acepts his help cooking

-they totally exchange recipes, and Leo gives her his number for any other mechanical misshaps

-"I work for good food and gossip"

-nona totally invites him again, and they totally become quick friends

-Leo tries pick-up lines on her and she's visceral with her criticism, but it sort of becomes their thing

-she does her best at learnign how to work with machines, and he's always ready to compliment and advice her

-Piper def notices and starts handing him her tasks, which he does bc he likes nona

-also he knows piper's been missing out on having free time for herself

-they gossip to each other, nona abt her book club or something, and Leo abt his coworkers and clients at his job

-Nona, an ally: why don't you have a partner? | Leo *bi disaster*: I'm too ugly

-nona is ready to throw hands with whoever hurt this boy

-Leo meets Hazel first, and the great grandparents thing goes both ways, and she has this old photo abt it

-it's a little weird but she's cool so Leo befriends her

-by the time Nico finds out, Leo's already besties with two out of three women in his life

-Nico totally thinks Leo's trying to get into Hazel's pants at first, even though Leo's nona's friend first

-Nico *protective brother mode*: wow, try hard must work so well for you | Leo *offended*: yeah, yeah, I'm too ugly to get any, can I fix your granma's ac now? | Nico: ?????

-you know that Nico thinks Leo's hot shit, like he thinks Hazel would date him, obvs he thinks he's attractive

-Nona totally smaks him when she finds out

-he finds out Leo meant the ugly thing, and is like, 'what the hell?'

-Nico: sorry I was an ass I thought you were trying to bang my sister | Leo: you thought I could bang your sister? have you seen her?? | Nico *gay disaster*: I've seen you!

-I woke up before there was any sort of clossure, so y'all'll have to figure it out for me

[runedwhip-archived]

— ANSWER THESE QUESTIONS THEN TAG 20 BLOGS YOU’D LIKE TO KNOW BETTER!

TAGGED BY. @merrywell​ TAGGING. whoever hasn’t been tagged yet!

NICKNAMES. izzie

ZODIAC. leo

HEIGHT. 5′2

TIME. 08:02 pm

FAVOURITE BAND / ARTIST. i always have such a hard time with this question, because it varies so much, and often times i usually listen to either movie scores or musical soundtracks. but some include ellie goulding, florence + the machine, panic! at the disco, fall out boy, imagine dragons, iron & wine, the civil wars

SONG STUCK IN MY HEAD. no ordinary girl from the h2o: just add water theme song lmao

LAST MOVIE I SAW. the holiday

LAST THING I GOOGLED. who to call if you find a cable down

OTHER BLOGS. i run a multi muse multi fandom blog at prophetavit, a very low activity piper halliwell over at witchfreeze, and of course this one. my multi is probably my most active blog at the moment, though as the holidays come to an end i’m working on bringing them all back to equal activity and having some sort of schedule so i can get things back active on all of them.

DO I GET ASKS. very, very rarely.

WHY DID I CHOOSE THIS USERNAME. izzy’s preferred weapon is her whip, and like to all shadowhunters, runes are important, soooooo.

FOLLOWING. 53

AVERAGE AMOUNT OF SLEEP. about 5 hours, give or take

WHAT I’M WEARING. basically pjs, i like to live in comfort

DREAM JOB. honestly? who knows. something with animals, something with books, something with coffee. something with travel, most specifically. i would love to be able to travel and write and be surrounded by animals. basically anything that lets me do that would be my dream job.

DREAM TRIP. currently i’d say europe and australia are high on my list! but really anywhere, i love traveling. i would love to go back and travel more of mexico again. and south america and really anywhere that’s in the coffee belt. currently i think my dream vacation would be starbucks’ origin tour to costa rica to our coffee farm there.

FAVOURITE FOOD. ooh, i love just about any mexican food, mediterranean food, italian, basically anything with lots of lemon and flavor!

PLAY ANY INSTRUMENTS. nope! i took unofficial guitar lessons in high school (as in, one of my classes was learnign the basics of guitar, but we didn’t go too in depth and only learned like, two songs maybe) and haven’t picked one up since. i like the idea of playing an instrument, but i think at this point in my life i don’t have the discipline to really commit to it. i would still be interested in learning the piano though, but dance is more where my artistic passion lays.

EYE COLOUR. hazel is probably the best way to describe them! they’re green on my birth certificate and basically any legal document, but there’s a lot of brown flecks in them that make them look more hazel colored.

HAIR COLOUR. red. think edward cullen hair, and that’s my hair color, though when i was younger it was basically a carrot orange.

LANGUAGES YOU SPEAK. english and spanish, both fluently. i can understand a fair amount of italian, but other than a few words, i unfortunately don’t speak it.

MOST ICONIC SONG. uh, honestly, i don’t even know haha.

RANDOM FACT. my entire work team consists of like, nine people? maybe ten, including my manager (it might still be nine, and eight of us plus her) and about over half of our team is composed of leo’s. it’s a good thing we all get along, because otherwise we’d be at each other’s throats all the time haha.

DESCRIBE YOURSELF AS AESTHETIC THINGS. i just realized i’ve done this before and i didn’t know how to answer this then and still don’t know either, so.

[itmrmeeseeks]

HW6 Case Q1

Link: https://qz.com/653575/can-we-trust-robots-to-make-moral-decisions/

Facts

1) There was revelation by Microsoft about the difficulty of creating moral robots. 

2) There was a Chatbot called Tay that was designed to speak like a teenage girl but the robot went wrong and turned into a Nazi-lover

3)There are two main approaches into creating an ethical robot

4) First is to make a specific ethical law for example to make maximize someone happiness, write the code for the law, and make a robot to strickly follow the law. 

5) Second is to create a machine-learnign robot and teach it how to respond to various situations to get at the satisfying ethical outcome.

6) Susan Anderson, philosophy professor at the University of Connecticut, and her husband Micheal Anderson, a computer science professor at the University of Hartford worked together to create robots that provide ethical care for elderly.

7) Anderson points out that in some ways, robots are superior ethical decision-makers to humans since humans are prone to making mistakes

8) She points out that the history ethics shows a steadily building agreement and work on robot ethics can contribute to refining moral reasoning.

Question:

What part of chatbot caused it become a Nazi-lover?

What are the disadvantages to the first method of creating ethical robots?

How long did it take the Andersons to finalize the “blueprint” of the creation of the ethical robot?

What challenges did they have with the creation?

As technology continues to grow, and robots to become able to make their own decisions, will it still be difficult to create ethical robots that won’t hack into computers?

[akiraaptx-blog]

The Boosting Approach

The origin of the boosting method for designing learnign machines is traced back to the work of Valiant and Kearns, who posed the question of whether a weak learning algorithm, meaning one that does slightly better than random guessing, can be boosted into a strong one with a good performance index. At the heart of such techniques lies the base leaner, which is a week one. Boosting consists of an…

[datascistuff-blog]

Generative Learning Algorithms: NAIVE BAYES

Generative Learning Algorithms: NAIVE BAYES

1. Input Data x is discrete. For example in text classification, x is a vector of 0 and 1 (bag of words). The vector size is the size of the vocabulary V.

2. In generate algorithms, we generate a model for the input data. For example, we can assume that the input data is Multivariate Gaussian Dist.

Q. What is the data input model in NB ?

  Its a MULTINOMIAL DIST. Since for a vector of size V, we have 2^V possible values of x drawn from a multinomial distribution. Since, we have 2^V possible values, defining or modeling x over this input space of 2^V-1 dim space is tough.

Q. How do we solve this dimensionality problem?

 By making the INDEPENDENCE assumption which is clearly not correct but works.

 What is independence ?? If x and y are independent, then

  p(x) = p(x|y)

in NB, this becomes

 p(x1,x2,…,xV|y) = p(x1|y) p(x2|y,x1)p(x3|y,x1,x2)…..

=p(x1|y)p(x2|y)…p(xV|y) 

 Words in a document are indepenent of each other.