wxs

If you like this you may enjoy the "Pink Trombone" an interactive speech synthesizer that you control by moving around the virtual tongue and lips: https://dood.al/pinktrombone/

> of consciousness, we have agents that perceive the environment and act, in order to maximize rewards

I don't believe "agents acting to maximize rewards" is a good description of any human I know, or if it is the reward function is certainly unknown.

Like Bret Victor? Who seems to have been involved in presenting this to Jobs? I'd say he's a pretty interesting guy and probably not worth firing even if you don't like one of his proposals.

https://twitter.com/worrydream/status/641705818585337856

I agree with the first half of this comment, but I don't follow you here: "Until there is an element of randomness fueled by context, they aren't creating anything new. That's where genius lies."

I suspect reframing of problems, seeing things in new light, paradigm shifts, new ontologies, whatever you want to call it are not quite so simple as context-dependent randomness! I don't think we understand this process very well right now.

Finally; I think there is an artfulness in copying existing patterns, because the way in which you "abduce"[1] an observation-explaining theory out of the infinite space of possibilities is a creative and aesthetic process.

[1] https://en.wikipedia.org/wiki/Abductive_reasoning

Unicode Technical Report #51, which is where Emoji are laid out, talks a bit about the current thinking of the committees on this:

> The longer-term goal for implementations should be to support embedded graphics, in addition to the emoji characters. Embedded graphics allow arbitrary emoji symbols, and are not dependent on additional Unicode encoding. Some examples of this are found in Skype and LINE—see the emoji press page for more examples.

> However, to be as effective and simple to use as emoji characters, a full solution requires significant infrastructure changes to allow simple, reliable input and transport of images (stickers) in texting, chat, mobile phones, email programs, virtual and mobile keyboards, and so on. (Even so, such images will never interchange in environments that only support plain text, such as email addresses.) Until that time, many implementations will need to use Unicode emoji instead

[1] http://unicode.org/reports/tr51/#Longer_Term

The matrix has nice features like flexible dates on a multi-leg trip that Google Flights doesn't offer.

I really like your Hidden Unit Zoo here http://colinmorris.github.io/rbm/zoo/ as a window into what this thing is actually "thinking" about. The "top matches" for a given hidden unit are pretty helpful.

The possibilities for how language will work in the future are really exciting and interesting! We made the Minuum keyboard, too, which also explores how machine learning assistance can open up new ways of communicating.

One reason we're interested in visual communication with Dango, though, is that regular text input is pretty good already. Chorded keyboards exist and are way faster, but people mostly can't be bothered to use them. QWERTY is just good enough. But the field is wide open for rich communication with images, nothing out there is particularly good yet.

Well those are sort of equivalent. But yeah, we use cross-entropy between the projected output and the target emoji distribution as our objective to minimize.

And yes, we do the t-SNE on that pre-projection space. That's why we can visualize the targets (emoji) in it. We can also t-SNE the word embeddings themselves — the input to the RNN — which is also kind of interesting. It automatically learns all kinds of structures there. Chris Olah has a good post on word embeddings if you're interested: http://colah.github.io/posts/2014-07-NLP-RNNs-Representation...

Not officially, although people have started reverse-engineering the website ;).

We can get an actually supported one up officially if there's interest. Email me at xavier@whirlscape.com!

Well you can search in Dango, too ;)

But yeah our main focus is suggestions. You can use Dango concurrently with the normal emoji keyboard, of course! It can just sit there showing you emoji you might not know about "ambiently"

Yeah we'd need to do some more work. But this is similar to stickers and GIFs: there are many fewer examples for any given sticker or GIF, which is why we do transfer-learning approaches as discussed in the article.

So there's a good chance we could get it to work! We've not focused on that possibility… yet.

Yeah so this is a legitimate concern. Of course sometimes it's fun to say "let's eat pizza :pizza_emoji:", but that's not hugely valuable.

However, Dango's training data includes people using Emoji to augment rather than repeat their sentence. So if there are two different interpretations and an emoji could disambiguate, the ideal is that Dango has seen people use that phrase both ways and, and that it suggests both possibilities and you can pick the one that you meant. In many cases this works now, in many cases we still have work to do.

It also suggests based on messages sent to you, so if there are a couple different replies it can show you them all (although this feature still needs work).

So yeah: we can focus on vectors at different levels of the net and these are in some sense different semantic spaces. In the article I talk about a level immediately before it projects onto the emoji vectors. If you look at the output after the projection (and do a softmax) you get a probability distribution across all emoji. This would be a different space in which each axis is an emoji, rather than the emoji being points distributed around the space.

Yeah this is a real challenge with AI systems, they reflect both our worst and our best back at us, we're still struggling with the best way to deal with this.

Although in this particular case it's actually just a bug: Dango gets confused by any skin tone modifier character, since they're not supported on Android (our target platform). Try putting in a "white" arm and you'll see the same results. They're actually just our "Dango is confused by this input" results.

We should fix the bug, of course!

We start supporting them as soon as they're available! Obviously we have a lot less training data early on but we can lean on some heuristics early on until it builds up.

Unfortunately in the app we can't give you emoji that your phone doesn't support so we don't always show all the results.

Thanks! I was tempted to go way more technical but we decided it would be fun to let more people get some intuitive grasp of these kinds of algorithms.

All over the web! For instance, over 20% of all tweets have emoji in them. Emoji are very popular in instagram comments, etc.

;)

Yes it can, language models can generate phrases that have not been seen before. Karpathy's RNN post is a pretty good intro to how that works http://karpathy.github.io/2015/05/21/rnn-effectiveness/

We use t-SNE for the projection! We do mention it somewhere in there. Could maybe be more clear

The sequences of emoji we ended up glossing over here (difficult balance making these concepts as accessible as possible). In the app we can beam search to predict combos, just as you would in sequence to sequence learning. That's not demo'd on the live website though.

I use Twitter on Firefox mobile and didn't know until this article that it's for some reason forcing an old ugly version. What UA string are you using? I tried the Android (Phone) one, and the Chrome one that the Phony extension provides but it still shows the ugly FF version of mobile twitter.

I went to the NYC incarnation a few months ago and it was pretty amazing, you can see some projects from that up here: https://news.ycombinator.com/newest

Glad we're getting our own version in Toronto.

His editor is very good, live preview of shaders, colour pickers, sliders for scalars, breakpoints. Check it out here: http://editor.thebookofshaders.com

That's no surprise: the event was hosted at ITP (http://tisch.nyu.edu/itp) which is a technological art program at NYU. Many of the participants were current students or researchers there.

Wouldn't the statement also be consistent with an increased percentage of car time with a passenger? That is, if all other costs were equal, and occupied time per driver-hour doubled despite per-ride profit halving?

Also see here for a number of sound samples of Whistled Turkish: http://www.newyorker.com/tech/elements/the-whistled-language...

I remember Adam from our batch! Cool kid: precocious, though he did come this close to spilling things on my computer a couple times :)

Though if I'm not mistaken the unprotected m4a still contains a "digital watermark" (i.e. metadata with your iTunes account info) which this article claims would become illegal to remove.

We use SQLite extensively in Minuum and I couldn't be happier. It's in all our critical real-time code paths that bring up candidates while you're typing. Because we're on SQLite our predictions keep getting faster for free!

Faster predictions mean we can handle more types of errors (more missed characters, more complex typos) in the same time budget. Thanks SQLite, you guys are awesome.