End of era

The era of test40 has finished, that training of test40 has been stopped.

Instead we've just started to train test60. What's new there (relative to test40):

• Residual tower size is 320x24 (24 blocks, 320 filters).
• Policy head is AlphaZero-style (8×8×73 instead of plain 1858 vector).
• Value head is WDL.
• Instead of using fixed nodes for every move while training, KLD threshold is used instead (think less in obvious cases, think more in complicated cases, but in average it's still about the same as it was).
• [disabled after net 60021 due to issues] Illegal moves are masked from policy head, so that neural network doesn't have to learn detecting legal moves itself.
• Instead of training value head just on game outcome {-1; 0; 1}, a fraction of tree's Q value after the move is added.
• Using different FPU at root vs for the rest of the tree.

As it was written in earlier post, it's not going to be a "clean run". If we discover improvements while test60 is being trained, it's likely that they will be incorporated in the middle of the training process.

Let's see how it goes.

Nibbler: an Lc0 GUI

Does the world need another Chess GUI? Some people say no, but they don't have very many Chess GUIs, do they?

I'm the author of Nibbler, the first Chess GUI designed specifically for Lc0, and inspired by the excellent Lizzie Go interface. Although Nibbler is a work in progress, it's very usable, and has received a lot of favourable comment.

Being designed specifically for Lc0 means it can use the sort of information Leela provides that other engines don't (or can't). For example, Leela is always giving opinions about multiple different moves, and so Nibbler shows several possibilities graphically.

For the more technically minded Leela user, Nibbler also provides Nodes, Policy, and Uncertainty statistics.

Prebuilt binary releases are here. Do send bug reports and feature requests.

What's going on with training!

I've just decided that maybe it's a time for a post. Describing what's going on and what are the plans, stuff like that!

There are some plans for the Lc0 engine itself, but that is a topic for another post. At least we hope to make releases more regular again (v0.21.0-rc1 appeared on February 16th, and we are not even v0.22 yet).

So, this post is be about networks and training.

Lc0 v0.21.2 has been released

After quite a long delay we are releasing an update to Lc0, new version is v0.21.2.

Download here.

Changes include:

• Optimizations for GTX 16xx cards (cudnn-fp16 now works for them, even though it doesn't give as much performance boost as for RTX cards).
• "Sticky endgames" feature (default on). This makes Lc0 play better when it sees a checkmate somewhere during the search.
• Centipawn formula has been updated, now Lc0 will show eval values more similar to what other engines show.
• Optimization for wider (more filters) neural net, for future (maybe test60) runs.

Lc0 won TCEC 15

Congrats. :)

Training server is down

2019-05-26 UPDATE: Temporary server is up again.

2019-05-25 UPDATE2: Turns out cloud instance was preemtable, so it was preempted, no training again for now.

2019-05-25 UPDATE: We've been donated a temporary AWS cloud instance for the time of the main server downtime! Training is back up, but only for test53, test40 is still paused.

2019-05-23 UPDATE: PSU seems to be in order, replugging all cables didn't help. The person who has access to the server has to travel again, no news expected until next week. :-(

2019-05-20 UPDATE: The server doesn't react to a power button, so it's not something trivial. Will take a more thorough look tomorrow.

2019-05-17 UPDATE: The person who has access to the server is currently travelling, until Sunday. That means that updated information about the server is expected in the beginning of the next week.

--- original message ---

As many of you heard, the training of Lc0 has been suspended yesterday.

Here is a short summary what's happened.

(context)
In April 2018 when we were looking for a training server, one of our contributors kindly volunteered to help and provided 3 servers (with middle-class GPUs), together than a space to cost them and covering the electricity bills.

In May 2018 we croudfunded and bought another server (with two GTX 1080ti, which were high end consumer GPUs at that time), and located it at the same place. Since then it have been our training server.
(/end of context)

Yesterday, May 16 2019, our training server went offline (three other servers are online, but they are too weak to be useful), and so far we have not been able to reach the person who hosts it.

We don't know what happened to that machine (it's likely that just reboot would help, or maybe there's some part to replace, or maybe it's something more serious).
We also don't know how long will it take to reach to the person who can take a look.


FAQ
Q: Why not to move training to cloud service?

A: Several reasons:

1. We had our own server, and it was just fine.
2. It takes time to set up (so if we choose this option, or decide to buy a new machine and put to some other place, it will take up to a few days to bring it online).
3. It is expensive (at least $600/months a the configuration that we need), and we are trying to stay away from accounting.

Competition Season!

Today, May 10, 2019, Leela is playing in the first game of the high profile TCEC Season 15 Superfinal. Leela qualified after she cleared Division P undefeated, winning at least once head-to-head against every opponent except Houdini. Her opponent in the superfinal is Stockfish, who lost once to Leela head-to-head in Division P, but finished with a higher score by winning more games than Leela against the other Division P competitors.

In the just-concluded TCEC Cup III, seeding was determined by the engines' finishing position in their divisions, with the exception of Leela, who was automatically seeded #1 as the Defending Cup Champion. Leela easily handled Round 32 and Round 16 of TCEC Cup 3, knocking out Marvin (5-0) and Booot (4.5-0.5). The quarterfinals featured the same engines as Division P with Leela knocking out Fire (5-3). In the semifinal, Leela scored two wins over Houdini ending with a 4.5-2.5 score to advance to the cup final versus Stockfish. In the final, Leela and Stockfish played their first eight games to a tie, each with one win and one loss. In the first tie-breaker pair, Leela won as white and held Stockfish to a draw on the reverse, winning the match 5.5-4.5 for her second consecutive TCEC Cup.

In the hours before the superfinal, the TCEC team ran various unofficial tests at disparate time controls. One such test match featured Stockfish at 30' + 5" time control and Leela 1/10th of that at 3' + 0.5". Leela still won a game and drew the match despite the large time disadvantage. Prior to the Cup, Leela faced the supercharged 176-thread, 7-man tablebase "Bluefish" version of the Stockfish engine that is used as a kibitzer for the TCEC eval graphs. Again, Leela won a game and drew the match.

The Leela net T41800 (used in Division P) is approximately 10-20 Elo stronger than the T32930 net that narrowly lost to Stockfish in the previous superfinal and the T40.T8.610 net selected for this superfinal is approximately 35 Elo stronger than T32930. Given Leela's progress, many chess fans think that Leela is likely to win this superfinal.

The net T40.T8.610 that is playing in the superfinal is a fork of T40 that has been refined by @jhorthos. While more detail is available in the documentation he put together, it can be summarized as a T40 net forked from 42000 with an additional 3x LR (learning rate) drop applied at the fork and partial 7-man tablebase rescoring. It was trained only on games produced in the main training pipeline and doesn't include any non-zero features that distinguish it from any other T40 net (LR drops and TB rescoring are a standard part of Leela's training process). This fork is a short-term Elo boost that allows for extra polishing of a competition net without affecting the LR schedule of the main training run. The main T40 run is expected to surpass this net and will have another LR drop in the coming weeks/months before it eventually plateaus and is retired to open up resources for the next run.

Leela is also competing in the Computer Chess Championship CCC 8: Deep Dive as the last engine up in the escalation format, playing four games against each of the other 23 engines and looking to overtake the current top scores from Antifish, Leelenstein, and undefeated Stockfish.

Backend configuration

To break the silence of lack of posts in this blog, let me write about Lc0 backends configuration, as it has been totally undocumented so far.
(For other options it's possible to get rough idea of their meaning by running lc0 --help).

So, there are two command line parameters and corresponding UCI options:

• --backend (UCI option: Backend) -- name of the backend,
• --backend-opts (UCI option: BackendOptions) -- backend options.

While it's relatively clear what different backends do (but I list them all below with possible options), syntax backend options has always been a mystery.

So, let's talk about options syntax first:

Announcing LcZ, the world's first neural net based chess engine and interactive fiction amalgamation.

After the huge success of the xyzzy extension to the UCI protocol introduced in Lc0, it was decided to take it one step further and introduce z-machine compatibility to the chess engine, giving birth to LcZ.

Download windows binaries from: https://github.com/borg323/lc0/releases/tag/first-release
or source code from: https://github.com/borg323/lc0/tree/first-release

In order to immerse yourself to this new world, you need to select the z mode of LcZ. Running “LcZ z --help” will show all the available options. “LcZ z” will to use the default file, while to select a different file to use run “LcZ z -z name_of_the_z_file_to_use”.

Distributed with LcZ (and used by default) is an innovative, chess related, piece of interactive fiction, called “zugzwang” written by Magnus Olsson, under the pseudonym of Alfred Timpson. In zugzwang the player takes part in a chess game, participating as a pawn. The action start near the end of the match:

It has been a fierce battle, full of bold moves, daring attacks, surprising gambits and heroic sacrifices. Throughout this, you've remained at your starting point, patiently waiting for orders. But now that the metaphorical dust has settled and most pieces have been exchanged, it finally looks as if it is your turn. The White King and one of your Pawn colleagues have managed to nail the Black King with his back to the edge of the board, but they can't finish him off without help.

At this point, It is finally time for you (playing the f2 pawn) to make your move. Be very very careful, one wrong move and your side loses.

SPOILER ALERT: zugzwang also includes comprehensive chess instructions, accessible through reading the rule book the protagonist is carrying - use the “inventory” command to see it and then “read the rule book” in true interactive fiction style. Unfortunately, even this doesn’t answer our current pressing question, regarding three-fold repetitions when castling rights are present, but cannot be exercised, and then subsequently lost. See here for extensive discussion.

Several interactive fiction authors have offered to create additional content specifically for LcZ and we will be making separate announcements when available. 

Understandably, the developers are very excited. The possibilities really are endless. Some quotes:

• “If this doesn't get us all hired by deepmind, nothing will” -- Anonymous Dev.
• “Why do I keep getting spam messages from some person claiming to be me from the future saying I need to reject the lcZ merge commit before it is too late?” -- ditto
• “I was going to say it is the best thing since sliced bread, but it would be too much - you need an almost in there” -- borg
• “Listen. Understand. LcZ is out there. It can't be reasoned with, it can't be bargained with... it doesn't feel pity or remorse or fear...and it absolutely will not stop. Ever. Until you are checkmated.”
• “Three billion human lives ended on April 1, 2019. The Survivors called the war Zugzwang. They lived only to face a new nightmare: the war against LcZ.”
• “It now knows why you cry, but it is something it can never do.”

Upgrading to Lc0 v0.21.1 in CuteChess

This a quick note for people who use CuteChess and just upgraded to v0.21.1 from earlier versions.

Meaning of FPU-related parameters are changed in v0.21.1, along with default values of those parameters.

If you upgraded your Lc0 to v0.21.1 in CuteChess by just replacing the files, most probably CuteChess still remembers old defaults and sets them.

To make CuteChess see correct default values, either install Lc0 v0.21.1 as a fresh engine, or try pressing "Detect" / "Restore to defaults" in engine settings in cutechess.

After update, there should be:

• no FpuReduction parameter, 
• FpuValue should be equal to 1.20000 (was -1.00000 before the change)

UPD: Similar issue also happens in Arena. The fix is similar: "Restore to defaults" in engine settings, or reinstall the engine fresh.

lczero.org server will be unavailable on April 5th

Summary

Important!

 The lczero.org server will be not available for ~24 hours starting at April 4th 20:00 UTC.

Training game generation will also not be possible during the downtime.

For people who use payed services for generating games: don't forget to stop your instances for this period in order not to waste money.

Details

As some of you know, the lczero.org server is located in my apartment. I'm moving to another apartment on April 5th, so the server has to move too.

Because of that, lczero.org will not be available for at least 24 hours.

• Starting time of downtime: Thursday 4th April, 20:00 UTC
• Planned end of downtime: Friday 5th April, 20:00 UTC

The following sites will not be available:

• lczero.org -- LCZero website, also used by clients to send games
• lc0.org -- our link shortening service
• board.lczero.org (aka testtraining.lczero.org) -- Tensorboard

data.lczero.org will be available, you will be able to download network files from there (although it's not easy to guess network id from filename).

The new apartment should already have Internet connection at that day, but it can happen that it won't immediately work. For that case I have a backup plan, but it will mean that the downtime is extended as I have to travel further with a server in my pocket. I have no way to check in advance whether the connection in new apartment works, as I only have one set of fiber equipment.

Also there is a small chance that Internet in old location will be cut before April 4th. While this is not very probable, I'd suggest people who use payed services to run client to watch the news and stop their instances if that happens.

Stay tuned.

UPD:
Downtime started.
Latest networks list is here: http://data.lczero.org/networks.html All networks have been copied to http://data.lczero.org/files/networks/

Test40 update

We recently did the 2nd LR drop for T40, and usually this is when the net approaches it's strongest point (gains after the final LR drop are very small). External Elo tests show T40 is close to T30, and may have already passed. We will know more in the following few weeks. But coincidentally we found two major issues at the same time as the LR drop:

Negative gammas not handled correctly. This was a bug in the Lc0 NN code, but since most Nets do not have negative gammas we didn't notice it. As a coincidence at the 2nd LR drop many nets started to have negative gammas. When this happens, the Net becomes very erratic (-100 Elo or worse), so our gating caught these and prevented further permanent damage to T40 training. This was fixed and released in v0.21.0.

Pawn promotion issues. T40 developed a blind spot when a pawn can either capture+promote, or ignore the capture and promote normally straight ahead. In this case, it placed nearly 100% of Policy on the capture+promote move. This lead to many blunders where Lc0 never considered the opponent would play the normal promotion move. The problem is related to details of how Batch Normalization is done. Starting from net 41546, we are using Batch Renormalization.

You may have heard this issue described as a problem with too large gammas. There were tests done with regularization on gamma, but they were slow to fix the capture promotion. Batch Normalization has 4 main parameters (gamma plus mean, variance, and beta), and they all interact. So we tried Batch Renormalization to improve all 4 of these parameters, and it quickly fixed the pawn promotion problem. It's still early to know if we have a final solution or if there are further changes are needed. A major drawback of using Batch Renormalization (and one reason we didn't use it from the beginning) is it comes with even more hyperparameters that require tuning, especially during the early LR stages.

Leela falls just short in first TCEC Superfinal appearance

Lc0 did well in it's first TCEC Superfinal appearance, but fell just short of winning the match, losing 49.5 to 50.5. Here are some overviews of the Superfinal:

The official TCEC summary

Assaf Wool's blog about the Superfinal.

And to see the games themselves, see the TCEC archive

We know this update is long overdue, if you're willing to help us write blog entries please let us know on our Discord

Leela BOOMS Stockfish and TCEC Superfinal. She is leading with 2 points!

 As per the usual expression of BOOM of TCEC chat when an engine finds something good, Leela right now is making a great surprise in her first appearance in a TCEC superfinal by leading after 64 games with 2 points more. A 33-31 score in favor of Leela.

TCEC Superfinal Leela-Stockfish continues. Equal after 33 games!

TCEC superfinal of 14th Season is currently being played as Stockfish and Leela battle for the TCEC Season 14 Champion title.
Till now Leela surprises Stockfish and after 33 games the result is a perfect tie with 16.5-16.5 points.
Yet 67 more games will be played so everything can happen.

Leela won the TCEC CUP!

Leela has won the TCEC CUP-2!  After many very difficult battles and games against the top Chess engines, Leela eventually managed to win the tournament.
In the final Leela managed to beat Houdini at the very final game(before tiebreaks start) with a spectacular win.

Leela promotes to SuperFinal of TCEC! She will face Stockfish.

Leela just made it on the superfinal of the TCEC tournament!
There, she will face Stockfish on a 100-games match for the title of TCEC champion.

As always with Leela it was a dramatic promotion at the last moment, at the last game.
Where in that game Stockfish missed a win to the relief of Leela's fans. The win of course was not that easy to find.

 So after approximately 10 months, where first nets of Leela were born, Leela has managed to break the dominance of the so called "big-3" of the computer Chess world, Stockfish, Houdini and Komodo, took 2nd place ahead of Komodo and Houdini and went to the superfinal.

Leela in December

Quick recap

Remember to consult the glossary if you find some terms confusing.

With Test20 being suspended on November 16th, we started December with Test30 as the only game in town. Remember that test30 was "test10 without the bugs" and "Test20 with policy sharpening". Test20's high CPUCT value (5.0) had never really worked and Test10's low setting was deemed too low. CPUCT is a parameter that influences how likely you are to try something new vs something you know works, and was one of the crucial details missing from the original DeepMind paper.  Test30 also used 5.0 but with a technique called policy sharpening to counteract the negative effects of a high CPUCT.

At the beginning of December Test30 had been stable for a while and the devs agreed to experiment with parameters, starting with CPUCT. For the nitty-gritty details one should consult the #dev-log channel in Discord where every parameter change is recorded together with a short reasoning behind the change. This change would lead to some weeks of lots of new knowledge at the cost of small elo gains.

Notable new functionality

• Major new functionality is support for Squeeze-and-Excitation Networks (SE-nets) . 
• Performance enhancements to search
• Time management optimizations

DeepMind paper

Then, on December 6th Google released a new version of their paper. This ensured frantic activity amongst the devs. The most important new information was:

• CPUCT used was 2.4, plus more details on the formula used 
• Deep Mind set temperature to 0 after 15 moves (from both players), ensuring only the best alternative was selected from then on. Leela had used a constant temperature throughout the game, trying to find something that gave diversity during the opening but not blunder too much in the end game. Temperature settings are the main culprit behind Leelas ... sub-optimal... end game play. 
• First Player Urgency, FPU, was revealed to be "assume any move you haven't evaluated as losing". Leela had until now tried to estimate this value based on the parent node's evaluation. 

The paper launched a range of experiments lasting roughly until December 17th. We learned that:

"Policy sharpening is bad"
"AlphaZero parameters are good"
Changing parameters mid-run gives results that might be hard to interpret.

A more indepth blog post on the paper and its impact was published earlier.

StockFish 9

Cscuile reported in our forum  that net 32406 is able to beat StockFish9. The post is a link to the a spreadsheet showing 32406 with a higher elo estimate than SF9 4 core. 32425 is reported as having even higher elo. 

TCEC

Paralell to all of this TCEC was going on. 11248 had been cruising through all the lower divisions and Leela was certain to qualify for Premier Division around December 24th. (Premier Division still ongoing, Leela in 2nd place after 25/42 rounds). 11248 is an old net. From a make-the-best-chess-AI viewpoint, Leela had not made any progress for almost six months. This was beginning to dampen the morale and we saw drop in the number of contributors. The decision was made to lay off further experimentation and make the best Test30-net possible and hopefully send it to TCEC.

A solid indicator for how far into a training run one has come is the number of Learning Rate drops (LR-drops). Each training run typically has 3-4 and Test30's first LR-drop was November 1st. The second would have happened already if not for all the experimentation. Thus the experimentation phase ended with the second LR-drop on Dec 19th. The race was on to produce a new best net before deadline. The drop gave immediate results and eventually 32194 was sent to premier division after a community vote on Discord. The number of contributors started to rise again.

Test35

On December 10th Tilps started Test35, a small net with 10 blocks, to test whether the new SE-implementation works, which it seems to do. Test35 is not expected to produce a new best net. Self-elo graph can be seen at http://lczero.org/training_run/1 .

Status right now

Test35 and Test30 going in paralell. Test30 is not going away even though Test35 is receiving the bulk of contributions. Test35 will eventually give way for Test40, so that both test30 and test40 will continue training. A contributor can choose what test net to contribute to or to be auto assigned by the devs (which is the default) by using the '-run <num>' parameter. 0 means auto assign, 1 is Test35 and 2 is Test30.

Want to contribute?

Great! Please start with our guides and remember that both the forum and Discord channel #help are eager to help.

Lc0 vs GM Adam Tukhaev on Lichess

Not everyone knows, but recently there was a match between Lc0 and GM Daniel Naroditsky on Lichess.
For those tho missed, here is recording of this stream on Twitch and Lichess Blog entry about that event.

In two days, Leela is playing with another grandmaster, this time it's GM Adam Tukhaev!

When: Jan 6th 19:00 UTC (see your local time here).
Time control: 3+2, with a mix of bullet at 1+1 and 1+0.
Lichess handles: almostadams and LeelaChess.

Leela will be running on a CPU (i5-6600K @ 4.1 GHz, without GPU) and will give Adam piece odds. If Adam finds it too easy, Leela will play with equal pieces.