issues
search
MTP-2023
/
mtp2023
1
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
train agent as both players (switch during training)
#63
tnowak1502
closed
12 months ago
0
agent as player 2
#62
tnowak1502
closed
12 months ago
0
multiplayer with winrate reward
#61
tnowak1502
closed
1 year ago
0
change challenges to 1 marble per player only
#60
tnowak1502
closed
1 year ago
0
implement agent evaluation (winrate vs other agents)
#59
tnowak1502
closed
1 year ago
0
implement agent deployment (vs human)
#58
tnowak1502
closed
12 months ago
0
implement self-training when agent beats mcts consistently
#57
tnowak1502
closed
12 months ago
0
verify mcts is working correctly
#56
tnowak1502
closed
1 year ago
0
adapt frontend for multiplayer
#55
tnowak1502
closed
12 months ago
0
git cleanup/restructuring
#54
tnowak1502
closed
1 year ago
0
server architecture
#53
tnowak1502
closed
12 months ago
0
roadmap slides
#52
tnowak1502
closed
1 year ago
1
adapt challenge generation for multiplayer
#51
tnowak1502
closed
1 year ago
0
adapt game logic for multiplayer
#50
tnowak1502
closed
1 year ago
0
adapt game env for multiplayer
#49
tnowak1502
closed
12 months ago
0
frontend mockups
#48
tnowak1502
closed
12 months ago
0
main into online learning
#47
tnowak1502
closed
1 year ago
0
Second Half: Project
#46
RezaXsys
closed
12 months ago
1
custom iteration statistic to track solverate during training (if possible)
#45
tnowak1502
closed
12 months ago
0
curriculum threshold as array
#44
tnowak1502
closed
12 months ago
0
implement online learning
#43
tnowak1502
closed
12 months ago
0
GPU Integration for Cluster runs
#42
tnowak1502
closed
1 year ago
1
Harder train set (more marbles)
#41
tnowak1502
closed
12 months ago
0
Implement curriculum mode where levels are combined
#40
tnowak1502
closed
12 months ago
0
Debug/improve AlphaZero agent
#39
mtemnov
closed
12 months ago
0
check if train and test set are disjunct
#38
tnowak1502
closed
1 year ago
0
Alphazero update (incl. other changes to agent application/training procedure/API)
#37
mtemnov
closed
1 year ago
0
Implement AlphaZero
#36
mtemnov
closed
1 year ago
2
Connect stored agent model to play games in the frontend
#35
mtemnov
closed
1 year ago
1
Setup of custom trainable and tune.run()
#34
mtemnov
closed
1 year ago
1
manual curriculum learning
#33
tnowak1502
closed
1 year ago
0
Wandb Integration
#32
mtemnov
closed
1 year ago
0
Implement playable challenges into frontend
#31
mtemnov
closed
1 year ago
2
Reward sync and new game variant
#30
mtemnov
closed
1 year ago
0
Sprint 3 Review - Open Points
#29
mtemnov
closed
1 year ago
1
Challenge Generator
#28
RezaXsys
closed
12 months ago
1
Code architecture
#27
mtemnov
closed
1 year ago
0
Create API endpoint to request start board and challenge
#26
mtemnov
closed
1 year ago
0
Evaluate challenge difficulty if baseline was adapted to require a specific switch positions
#25
mtemnov
closed
1 year ago
0
Synchronize reward function usage with baseline algorithms
#24
mtemnov
closed
1 year ago
1
Agent training for baseline game variant
#23
mtemnov
closed
1 year ago
0
React frontend update
#22
mtemnov
closed
1 year ago
0
Merge code from Cluj branches into main
#21
mtemnov
closed
1 year ago
0
Document fundamental project setup and code architecture
#20
mtemnov
closed
1 year ago
1
explanation video for training run
#19
tnowak1502
closed
1 year ago
1
React
#18
2XG-DEV
closed
1 year ago
1
test case jsons folder, pass to game env using config
#17
tnowak1502
closed
12 months ago
0
Merge hackathon results into main
#16
mtemnov
closed
1 year ago
1
Weights&Biases Integration
#15
mtemnov
closed
1 year ago
2
Reward Design Theories for RL
#14
RezaXsys
closed
12 months ago
0
Next