Matetricks

Hey—congrats on the launch! I'm a big Valorant player (peaked Immortal 3 a few seasons back) and I've also used/talked to the people at visor.gg and pursuit.gg when they were around.

Totally understand the concerns here around being at the whim of the publisher, and previous companies have been blindsided by changes in how companies like Blizzard decide what is and isn't allowed on their platforms. Interestingly, YC has had a few of these companies go through similar experiences but I think they still think there's something to be built in this space.

For what it's worth, I think this is a good idea that's going to run into similar issues, regardless of what safeguards are put in place to avoid stepping on the publishers' toes. I'm sure there are some learnings from talking to Ivan or James from the last two YC companies that tried to do this. Also happy to share some more of my thoughts :)

Congrats Geoff! You helped us a lot when we were young, first-time founders and we learned a lot from you and Tim. Thanks again for all your advice—excited about the future for YC :)

It was called Chesscademy and we were raising $750k.

We were tackling the adaptive learning problem by creating tailored curricula. The plan was to move outside of chess to other subjects as we grew, but we found monetization to be very difficult.

Chris is an amazing person who takes the time to get to know you.

When we were fundraising, instead of having us do a regular pitch he took me to an event where Peter Thiel was giving a talk about his new book. We were a chess education company, so he thought it would be good for us to talk with Thiel as he's a chess master himself.

He didn't end up investing in us but my experience with him was much more memorable than any of the other meetings we had while fundraising.

How does this compare with other companies in the space like Chorus, Gong, and other YC companies like People.ai or VOIQ?

Congrats on the launch!

I think the move rate is capped at two per second, so even bots will have limitations in their movement speed.

The list that's returned still contains mainly tactical surprises where Stockfish inaccurately evaluated the position at the end of depth 5. I think what I'm trying to say is there are some moves in a position that aren't tactically surprising (a piece sacrifice, a crazy attacking move, etc.) but positionally surprising (a long maneuver to get a piece to a certain square that I didn't think of). These positionally surprising moves aren't captured by this methodology because they don't involve large fluctuations in valuation when the depth changes.

As to your second point, an issue with how computer chess affects the modern scene is how playing the "best" move in any given position isn't representative of how humans play. Humans carry out plans and evaluate positions to the best of their ability, but the heuristics and procedure they use aren't the same as a computer's. For example, Karjakin didn't prepare for his match against Carlsen last month by playing a bunch of games against Stockfish. Rather he probably analyzed Carlsen's past games and opening choices to come up with a strategy.

I do think you can come up with a way to prepare against individually known opponents by identifying weaknesses programmatically. You can model a human's approach to playing chess as a distribution of parameters (material, king safety, pawn structure, etc.) that take in the current position and return the best move. You also have Stockfish's evaluation which returns the "best" move. With this, it's possible that you could use build a neural network that learns to play very similarly to a certain player by using their past games as a training set and comparing the chosen move to Stockfish's move. The network could learn to mimic the heuristics that the human individual uses to make decisions and playing against this new AI would be great practice for preparing against specific opponents.

If I understand your code correctly in analyse_evaluations, you're defining a "surprising move" as a move that has a large change in valuation when it's considered at a higher depth. So if a "human" (really Stockfish at depth 5) evaluates a move as +1 and a "computer" (Stockfish now at depth 11) evaluates the move as +5, the move is surprising.

This is pretty interesting, but I'm not sure if it fully captures all the nuances of what a surprising move is. You might be able to classify a move as tactically surprising if it becomes clear after depth 7 that the ending position is favorable. However, in my opinion truly surprising moves are ones that carry plans that I haven't even considered. Hence, this methodology doesn't capture moves that are positionally surprising as there wouldn't be such a drastic change in evaluation at different depths. I'm not sure where you would start to figure that one out though :)

That being said this is really cool work!

I didn't build this, it's a site that's been trending on my news feed. Two people in my network built it over a weekend.

They're probably updating the servers so that the games won't be as laggy. Shouldn't be long!

They've added a few more ML courses—the go to class for undergrads is now ORF 350 (Analysis of Big Data) with Han Liu. ELE 535 (Pattern Recognition and Machine Learning) is also a new course that was added last year.

For those interested in the "Probability and Statistics" portion of the guide, Princeton recently created a certificate program in statistics and machine learning that has some more updated information on courses: http://sml.princeton.edu/

I'm pursuing the certificate right now and the courses have been great so far. Princeton's known for having a rather theory-heavy approach in their quantitative classes but I've found a good balance with applications in some of the classes (COS 424, COS 402).

I saw their presentation over the summer and was super impressed. I didn't participate in the initial challenge but I'm excited to start now!

Have you heart of StackSort?

https://gkoberger.github.io/stacksort/

Congrats to all the IK12 folks! Can't wait to see more of your amazing work as part of YC.

It's worth mentioning that Lee Sedol mentioned in an interview that even if he loses a single game against AlphaGo, he will have lost the match. He was expecting to win all 5 games.

Congrats on all of your progress! I remember being one of the first masters to use Lichess and I absolutely loved it.

Currently, you can play through the solutions using the left and right arrow buttons underneath the board.

Thanks - we've realized that there are some problems with FB connect when characters like a period are used. We're working on fixing this.

Bringing in Stockfish is definitely on the road map - this will allow us to automate some of the checking processes.

I completely understand where you're coming from - without feedback telling you why the move you chose is incorrect, the exercises can become a bit frustrating.

We're aiming to make the site far more adaptive by tailoring error responses to your moves. At the moment, I'm going through the tactics in Train and tagging them with motifs and supplying explanations for the solutions. An idea is to track the most common responses and create specific messages to be shown when those moves are made.

We are using chessboard.js (http://chessboardjs.com/) for our board right now and I'm not sure if they support two-click yet. We may have to add it ourselves. Thanks for the note!

We're building an onboarding process for tactics so the objective is clearer, but the point of the section is to put you in a position that you might encounter over the board and have you find the optimal moves.

If black captures the hanging pawn, that is known as the Open Variation of the Ruy Lopez. I didn't want to get too deep into theory with this video as it was only supposed to outline opening ideas. White regains the pawn back due to his pressure on the e-file.

You can read more about the variation here: http://en.wikipedia.org/wiki/Ruy_Lopez#Open_Defence:_4.Ba4_N...

We definitely don't want our users to be getting the solutions via brute-force. We're considering allowing the user to give up after a certain number of failed solutions. Incorporating more specific feedback to the user's incorrect moves would also help with this problem.

Are you referring to the structure of the content on the site as a framework? We think that this approach to education has significant applications beyond chess, but at the moment we're solely focused on making Chesscademy work for its intended purpose.

We pulled them from a database of puzzles, so there probably are a few bugged ones here and there. We've pruned the majority of tactics that are simply broken, but I'm sure that a few questionable ones remain.

I'm in the process right now of going through all of the tactics and adding motif tags along with explanations for the solutions. While I'm doing this, I'll be sure to make sure that all the tactics make sense. Thank you for your feedback!

Thanks so much for the kind words! Happy learning!

Thanks for your comment! Please email me at andrew@chesscademy.com with any feedback you have for the site.

What distinguishes Chesscademy is our focus on education. There are a lot of sites that offer a variety of features - multiplayer, tactics, etc. However, no other service offers educational content structured in a way that provides users with a feeling of progression. We aim to incorporate qualitative feedback into every aspect of the site so that your specific weaknesses can be tackled.

Chesscademy is also aiming towards the edtech market - schools have expressed interest in using our service in the classroom as part of their chess program. Chesscademy works both in and out of the classroom and complements instructors of all levels. For example, teachers can assign students courses and track their progress over time, using these data points to address particular topics.