Log in

No account? Create an account
< back | 0 - 10 |  
Templar [userpic]

Some afterthoughts on 4th Kyiv's intellectual "What? Where? When?" marathon.

July 31st, 2017 (12:44 pm)

1. ~20 hours of the nonstop quiz is hardcore! Was, is and will be.
2. Still absolutely worth doing for many reasons. The simple feeling of achievement is enough, it's hard to explain how satisfying the "I can do it! I made it!" feeling. Also, it's one of the safest ways to put your mind in altered state. After sleep deprivation and heavy randomized load for multiple hours, you can actually feel how different your brain works now. And, as someone already noticed in comments to the event, there are some works that suggest it helps with anxiousness. I can confirm - for me, it felt like break statement inside the infinite loop. Some thoughts that weren't pleasant but was hard to get rid of, just stopped dead in their tracks.
3. The increase of distance from usual 36-48 to 500 questions shifts perspective and priorities. Missed 5-10 questions become a much smaller problem. One of the funny side-effects - no WC lines ever, even though we had a ton of people there. During small tournaments, everyone tries to relieve themselves during short breaks, and here - well, whenever you need this you just go and return few questions later.
4. My effective distance is around ~250 questions right now. Next time I need to plan my break to have ~50 questions rest after 250 played. That should help increase team results. After that mind enters some kind of limbo state - you aren't sleeping, yet it's hard to get all information in the same place and comprehend question as a whole - words just scatter in your head. It's especially hard in "W?W?W?" because most of the questions are built around complex interactions between facts, not singular facts themselves. On the other hand - when you try to focus really hard in that state - your brain tries to utilize some unexpected resources, like visual parts - some of the questions actually became mini-dreams with the vivid visual component.
5. It's hilarious when you actually manage to find the answer in that state.

Me, awakening from half-sleep state and looking at city emblem in the handouts, while hearing last words of the questions "... and that is the proof that THEY will win": "Heeey, that really looks like Tyrrels emblem from GoT. Maybe they will win the game of thrones?". Answer: Tyrrels! And all this time handout was upside-down.
6. Many thanks to amazing people in my team for a great time! You're the best!

Templar [userpic]

О воспитательной роли сказок

July 5th, 2017 (10:57 am)

Обзор сказки о том, что нужно даже в самом страшном чудовище видеть хорошее и его тоже можно любить.
Критикуют что там появляются намеки на другую ориентацию и люди немного не того оттенка кожи.

Templar [userpic]

One-Punch Man 075

May 28th, 2017 (11:59 pm)

Observation - even those few who (like Genos) understand that Saitama indeed insanely strong - still underestimate him. It seems that power gap between him and any other hero quite literally beyond comprehension.

Templar [userpic]

Idea: mind as a tapestry (Or a choir, if you like that more).

May 24th, 2017 (01:55 pm)

I know that I already read somewhere about a similar idea, so it's not new, just playing around with it.

Everyone that had an inner argument (and AFAIK that means literally - everyone) know that you don't always act as a unit and want one thing. More often you have a set of probably self-contradictory wishes/desires/obligations and when you actually decide what to do you need to balance them out, reach consensus or at least compromise between all of them.

From the architectural point of view that makes sense - neural networks have self-similarity property, so the only difference between a part of the neural network and the entire thing is its size.

Note: I know that brain is more complicated, and there are some parts not similar to others. This is a (probably over-)simplification.

So what we consider as our internal monolog can be just emergent behavior created by smaller sub-minds taking sides, arguing with each other and trying to reach a compromise.

Linked idea-1: On the importance of stories (regardless of medium - books, movies, games...).

What we observe in stories are patterns of behavior and thinking in situations we never experienced ourselves. Every character, every decision made can influence you - you take those parts that you find inspirational, and you add new patterns to your neural network tapestry, new voices into your inner choir. So when your commander Shepard decides to stand up against all the odds - you literally add small "Commander Shepard"-like voice into your head that will tell you to do the same next time YOU will face a similar situation. Of course, that doesn't mean that you will do the same, but there will be +1 vote for that option.

"Empty" stories are just stories that don't add anything new - cliche characters and same overused story archetypes that are already there. So reading them is easy but you can't quite remember them in a week - they didn't leave any lasting change in yourself. Probably some internal links were reinforced a bit, but that's all.

"Bad" controversial stories are just dissonance - it's not like there is nothing to add, but what is already there in your head don't fit with these new patterns, the result just looks worse than it was before. So you try to repel it. And unless this story is something crucial/useful - you probably don't need to force yourself.

"Good" story that makes an impact and you can remember it for years is something that makes you richer, increase internal harmony.

Linked idea-2: Psychological problems as sub-minds clustering, disharmony between internal patterns.
Sometimes you just can't reach a compromise inside yourself. Sometimes there is "splinter cell" inside your brain, some voices that just tell you to do something very different from what all others say. When it's small - you may still have one main personality, but you always feel anxiousness, like some part of you consistently disagrees with what you do. OCD rituals may be the way to either isolate that splinter cell (ritual cause neural path "around" it to fire and block messages) or silence it (maybe that splinter cell tells you to do A, B, C and wash your hands, so by washing your hands you do part of what it wants, and all other points can be ignored for a while). And when that splinter cell is comparable in "size" to other - you have multiple personality disorder. So instead of just one emergent identity - you have two (or more) completely different, each with its own set of voices behind it.

Linked idea-3: And now let's go crazy and think about Tolkien and it's creation myth.
What if "music of the Ainur" is actually a metaphor for new personality in making, a literal birth of a character. New voices adding to the choir in harmony, little splinter cell started from Melkor, Valar as chief sub-minds that are trying to fight with it, but as all internal arguments always do - leaving "Middle-earth" in ruined state...
Fun stuff, I can actually see this as a thesis.

Templar [userpic]

Idea: support vectors in AGI alignment problem.

May 24th, 2017 (01:54 pm)

(I need to mention that I am in no way AGI specialist, so maybe this will be quite naive...)

Usually when you hear/read about AGI-A problem, examples of what they are trying to achieve and what is considered for "training" are quite obvious and extreme - "killing people is bad, inventing disease cures is good" style. While this will be certainly useful for unit testing already aligned AGI, I don't think it will be so useful for training. Consider SVM type of algorithms that used for classification problems - what there called "support vectors" aren't extremes of their classes, quite contrary - they are most similar examples. You want your algorithm to learn subtle differences between very similar outcomes, not broad categories with 99% in "gray zone."

In a similar fashion you want to train AGI on controversial tasks, (in)famous "trolley problem" type and so on, specifying what your particular AI should consider the right choice.

Of course, this opens an entirely new can of worms, because humans themselves can't agree on what is the right thing in those problems, but this will allow more fine tuning of result AGI developers trying to achieve.

Templar [userpic]

Я вернулся!

May 24th, 2017 (01:53 pm)

У меня в кои-то веки появилось желание вылить мысли на бумагу.
Не знаю как долго оно продержится, но пока что я снова здесь.
Не то, чтобы я когда-то пропадал - я всегда читал ленту :) просто писать - это обычно не моё...

Не удивляйтесь постам на английском - некоторые мои знакомые англоязычные, да и большинство моих друзей этот язык знают.

Templar [userpic]

(no subject)

August 3rd, 2016 (02:47 am)

Мама умерла. 2.8.2016. вечная память....

Templar [userpic]

С праздником всех!

December 31st, 2014 (09:42 am)

Счастливого Нового года вам, пусть все будет чудесно и волшебно для вас в нем!

Templar [userpic]

C Новым годом, друзья!

December 31st, 2013 (12:10 pm)

Всем счастья и волшебства в этом новом году ^__^ пусть он принесет вам много-много радости, а также маленьких и больших чудес в жизни!

Templar [userpic]

книжный флешмоб "семь смертных грехов" от Ровенны

May 20th, 2013 (02:32 pm)

1. Жадность. Ваша самая дорогая книга?
Хм... Наверное "Всеобщая история дипломатии" в подарочном издании - приз за одно из выступлений ЧГК.
Или "Хроники Амбера" - антология одним томом.

2. Гнев. Какое произведение вы бы сожгли?
Никакое )) все могут служить примером - кто положительным, кто отрицательным. Пример "как и о чем писать НЕ стоит" может быть весьма полезен.

3. Обжорство. Какую книгу вы с удовольствием заново и заново перечитываете?
Хроники Амбера )) Очень трудно оторваться, сколько бы раз не открывал. Но в мире столько новых книг ^^ очень редко когда не жалко времени на то, что уже читал.
Еще сборник стихов "толкователь снов" :)

4. Лень. Какую книгу вы бросили читать?
"100 лет одиночества". Стыдно, потому что обещал очень прочесть, но в то же время не жалко - стиль совершенно не для меня, И шел туго и что я мог бы из этой книги извлечь - тоже непонятно.

5. Гордость. Какую книгу вы упоминаете, если хотите выглядеть интеллектуалом?
Одну книгу? Интеллектуалом? Да вы верно шутите ))))
Разговор, в котором не было упомянуто минимум 2-3 - просто пустой и можно сказать что его и не было.

6. Похоть. С каким литературным персонажем вы хотели бы переспать?
Хммм... никогда не думал в таком ключе о героинях книг )))

7. Зависть. О какой книге жалеете, что ее написали не вы?
"Толкователь снов" упомянутый выше :) хотел бы я так уметь передавать сказку и волшебство словами...

< back | 0 - 10 |