Live data from Hacker News

Generating Magic cards using deep, recursive neural networks

mtgsalvation.com

21–30 of 69 posts

Re: Generating Magic cards using deep, recursive neural networks

#21

> Shring the Artist > 2BB > Legendary Creature - Cat > Flying > Whenever you cast a spell, you may return target creature card from your graveyard to your hand. > 2/2 That's actually a fairly plausible card, in terms of mechanics. It's thematically appropriate for black, it's about the right cost, it's powerful but specialized (returns creature cards only, and only to hand rather than to play), and it's legendary to…

> The only thing slightly odd is the flavor: it's a flying cat.

Flying cats are actually already in the game. Back in the Planar Chaos days I used this one in a really trollsy black deck:

http://gatherer.wizards.com/Pages/Card/Details.aspx?name=mir...

Re: Generating Magic cards using deep, recursive neural networks

#22

> Shring the Artist > 2BB > Legendary Creature - Cat > Flying > Whenever you cast a spell, you may return target creature card from your graveyard to your hand. > 2/2 That's actually a fairly plausible card, in terms of mechanics. It's thematically appropriate for black, it's about the right cost, it's powerful but specialized (returns creature cards only, and only to hand rather than to play), and it's legendary to…

I'd agree except for the legendary part, there's really no mechanical reason for it to be legendary, the effect isn't particularly powerful.

Re: Generating Magic cards using deep, recursive neural networks

#23
post #21

> Shring the Artist > 2BB > Legendary Creature - Cat > Flying > Whenever you cast a spell, you may return target creature card from your graveyard to your hand. > 2/2 That's actually a fairly plausible card, in terms of mechanics. It's thematically appropriate for black, it's about the right cost, it's powerful but specialized (returns creature cards only, and only to hand rather than to play), and it's legendary to…

> The only thing slightly odd is the flavor: it's a flying cat. Flying cats are actually already in the game. Back in the Planar Chaos days I used this one in a really trollsy black deck: http://gatherer.wizards.com/Pages/Card/Details.aspx?name=mir...

It's a vampire, seems slightly more legitimate.

Re: Generating Magic cards using deep, recursive neural networks

#24
post #23
post #21

Earlier quoted context omitted.

> The only thing slightly odd is the flavor: it's a flying cat. Flying cats are actually already in the game. Back in the Planar Chaos days I used this one in a really trollsy black deck: http://gatherer.wizards.com/Pages/Card/Details.aspx?name=mir...

It's a vampire, seems slightly more legitimate.

There are others, mostly cats riding stuff that flies.

http://gatherer.wizards.com/Pages/Search/Default.aspx?action...

Re: Generating Magic cards using deep, recursive neural networks

#26
Wow, this has to be my favorite HN post. I couldn't finish reading it at work due to tear filled laughter.

I think it is less funny if you don't know anything about MTG. Which is interesting - if you don't, or don't know it well, these cards seem fairly indistinguishable from real cards. The differences are hillarious.

Mointainspalk.

Re: Generating Magic cards using deep, recursive neural networks

#27
post #17

Magic is an interesting game just from an text analytics perspective. Using a source like mtgjson you get a reasonably sized corpus of text that looks natural enough for a human to comfortably grok but sufficiently templated to apply even naive methods and acquire awesome results. I built a tool that uses n-grams to find cards similar to each other. The result is comparable to manual suggestions done by users but tak…

Is this tool open source anywhere?

Re: Generating Magic cards using deep, recursive neural networks

#28
I was very excited to learn about recursive neural networks and how they are different from recurrent neural networks. I began imagining self-similar fractal topologies and automatic convolution layer creation etc... Disappointed to see it was just a typo :-)

Re: Generating Magic cards using deep, recursive neural networks

#29
post #8

How about IDE that suggest you next line of code? Or next variable, or method to call? Add static analysis and it can actually be fairly useful. E.g. val name = object.[maybe you wwant to use getFirstName]?

If a prediction engine can consistently predict your next line of code, wouldn't that be a sign that you're doing way too much copying and pasting instead of abstracting common patterns?

Re: Generating Magic cards using deep, recursive neural networks

#30

> Shring the Artist > 2BB > Legendary Creature - Cat > Flying > Whenever you cast a spell, you may return target creature card from your graveyard to your hand. > 2/2 That's actually a fairly plausible card, in terms of mechanics. It's thematically appropriate for black, it's about the right cost, it's powerful but specialized (returns creature cards only, and only to hand rather than to play), and it's legendary to…

I'd agree except for the legendary part, there's really no mechanical reason for it to be legendary, the effect isn't particularly powerful.

Duplicate it a few times over via any number of duplication mechanisms (common enough in red or blue), and playing one spell would get you multiple creatures back. Since those creatures themselves are spells, it'd be too easy to set up a loop or combo. Making it legendary means you can't easily end up with "play one spell, get back N creatures".

For that matter, creatures are cards, and there are effects powered by discarding cards or cards going to the graveyard, so if you can play one spell to get 5 more cards in your hand, and then discard 5 cards to get an effect...

Post reply on HN