List your company

ResearchFormerly WizWand

cartpole

Benchmarks

TaskDatasetSOTA resultTrendResults

Neural Network Verification

cartpole

142Time (s)

3

Reinforcement Learning

CartPole

2,000,000Total Episodes

3

Meta-Reinforcement Learning

Cartpole fl-ood

0.004FLOPs (k)

3

Continuous Control

cartpoleSerialTriple

0.736Average Reward

3

Continuous Control

cartpoleSerialDouble

0.272Average Reward

3

Continuous Control

cartpole ParallelDouble

0.549Average Reward

3

Continuous Control

cartpoleBalance

0.951Average Reward

3

Continuous Control

cartpole

0.844Avg Reward

3

Safe Exploration

Cartpole (eval)

46.4Min Filter Rate

2

Safe Exploration

Cartpole (train)

382Total Infeasibilities

2

Trajectory outcome classification

CartPole LQR

86Precision

2

Closed-loop verification

Cartpole

100TNR

2

Reinforcement Learning

CartPole standard (test)

3Length

2

Interpretability Evaluation

CartPole

5Interpretability Score

2

Adaptation to Non-Stationarity

Dynamic CartPole

45Recovery Time (Episodes)

1

Meta-Reinforcement Learning

CartPole

—Primary metric

0