no apologies for the title

Reinforcement Learning (RL) follows a simple premise: there is a world, which you can sense and interact with, and a measure of goodness or badness which you can influence through your actions. Everything else is variations on trial and error. The actual algorithms are intuitive and often provably optimal, given their assumptions. Everything about the complexity comes from determining constraints, sensors, actions and values — all setup before the actual computer science begins. A straightforward example would be a chess board — the state of the world is the arrangement of pieces on the board, and we can influence the…


and do your words reach heaven’s ear?

I recommend you first read this article I wrote on word2vec, if you haven’t already. If you have, go back and reread it anyway because it’s been a year and a half and I don’t trust your memory. Here’s the two sentence recap: words can be mapped to 50-dimensional vectors which represent their location in some high dimensional “semantic space”. This isn’t just a quirk of computers, these numerical representations actually map to some inherent structure in language, and vector math applied to these words gives meaningful results, the prototypical example being that V(‘king’) — V(‘man’) + V(‘woman’) = V(‘queen’).


what is left

They knew damn well what they were doing. They knew it was frivolous, unnecessary. Other places haven’t done this, and they’re still better off than we are. I don’t mind for myself, so much, even though it hurt me, too. But I go out there and tutor these kids — I come into their homes, sit down at their kitchen tables and watch their faces go blank with nihilistic despair as they open the laptop. This website hosts worksheets, this other website is where you turn them in. Lectures? Office hours? No, nothing like that — just poorly phrased multiple…


and uh this indicates the employee’s ahm current location

Peter Wilksy sat at his desk and tried to focus on the report in front of him. His third cup of coffee was doing little to help him stay awake, and he found his mind drifting again to Jenny, who sat on the other side of the cubicle cluster and two spots to the left of him. She wasn’t the hottest person on their floor, but she was the nearest hot person to him, and he found himself wondering what she was doing at that moment. Maybe frowning at the screen, bent forward slightly, pouring over the rows and columns…


how could this happen to me

There’s a lot of to-do about radical-ization online right now, about the online right. Many people complain about things like Youtube “algorithms” “feeding” viewers “radicalizing” “content” from “gateway” sources, (can you sense my “contempt” for this “shit” yet) and wireheads share stories about their best buds who used to be great progressive guys, but watched one too many Ben Shapiro videos and mutated into horrible slavering werewolves, with s0nnenrad tattoos and jean vests.

The outrage over this issue is simultaneously under- and over-blown. Let’s start with the surface level first: there is no algorithm designed to move you from Ben…


Dr. Ramesh ran the script again.

It was his third week wrestling with this script, but his 22nd month working on this problem — thousands of pages of reading, two notebooks full of handwritten notes and half-finished proofs, several different implementations, collaborations with multiple colleagues, including interdisciplinary grants, and one marriage: that was the cost he had sunk into trying to isolate this pattern to date. He had stopped considering that cost, however. …


when I was a kid we had a wood burning stove instead of a heating unit

When I was 14, I got in trouble. I don’t remember what for, I think it was lying, it doesn’t matter. The point is it was November, and there was a half cord of wood on the back porch. It was decided that for my punishment I would shift the wood from the porch to out by the shed, and I would cover it with a tarp for the winter. It was near freezing, so I put on my coat, a wool cap and heavy winter boots, and walked outside into the gray afternoon to pay my debts to society.


muh

“Freedom of speech” is such a limited thing as to be almost worthless. Every line of communication between you and the general public, and most lines of communication between you and your close friends, are mediated by a third party. If you post on Twitter you’re subject to Twitter censorship — if you make an independent website you can get deranked by Google. If you spread your website via word of mouth it’s still subject to Cloudfare’s approval, and even if you host independently CenturyLink can deny your servers. You can have videos pulled off the air, books taken off…


five nines

Creaking noises above my head. I can’t focus, can’t concentrate, I have to get out. A dog howls in the yard outside my window. My hand shakes as I grab for the kettle, and I almost spill boiling water all over my feet. It has been one and a half days since I’ve talked to anyone, but I’ve been reminded of other people’s existence every waking second. The air ducts hum quietly, perched on the edge of my consciousness. My next door neighbor is watching TV — sports, probably. I can hear the professional tones of a commentator’s voice even…


the world is not enough

Here’s how it works — you walk in and ask for an application. You fill out a couple meaningless questions that nobody will ever read — the only thing that matters is the hours of the week you say you’re available for. You might think the number you put in for “Desired Wage” matters too, but you’ll find out differently at the interview, where the owner tells you matter-of-factly that you’re going to make eight bucks an hour doing this, but you might get a 50 cent raise next year if you do your job well. Your opportunities within the…

Ostav Nadezhdu

idiot

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store