API-documentation

• MMLF package interface
• Agents
  • Agent base-class
  • Actor-critic
  • Direct Policy Search (DPS)
  • Dyna TD
  • Fitted R-Max
  • Model-based Direct Policy Search (MBDPS)
  • Monte-Carlo
  • Option
  • Random
  • Temporal Difference Learning
    • TD(0)
    • TD(lambda) - Eligibility Traces
• Environments
  • Single-agent environment base-class
  • Fully-observable Double Pole Balancing
  • Linear Markov Chain
  • Maze 2D
  • Maze Cliff
  • Mountain Car
  • Partially-observable Double Pole Balancing
  • Pinball
  • Seventeen and Four
  • Single Pole Balancing
• Observables
• Resources
  • Function Approximators
    • Cerebellar Model Articulation Controller
    • K-Nearest Neighbors
    • Linear Combination
    • Multi-layer Perceptron (MLP)
    • Multilinear Grid
    • QCON
    • Radial Base Function
    • Tabular Storage
  • Learning Algorithms
    • Eligibility Traces
    • Temporal Difference
  • Models
    • Grid-based
    • K-Nearest Neighbors
    • Locally Weighted Projection Regression (LWPR)
    • R-Max Model Wrapper
    • Tabular Model
  • Optimization
    • CMA-ES
    • Evolution Strategy
    • Random Search
  • Planner
    • MBDPS Planner
    • Prioritized sweeping
    • Trajectory sampling
    • Value iteration
  • Policies
    • Linear Policy
    • Multi-layer Perceptron Policy
    • Value Function Policy
  • Policy Search Methods
    • Fixed Parametrization
  • Skill Discovery
    • Predefined Skills
    • Local Graph Partitioning
• Framework
  • World
  • Spaces
  • State
  • Monitor
  • Experiment
  • Filesystem
  • Interaction Server
  • Protocol