Planning with an Adaptive World Model

We present a new connectionist planning method TML90 . By interaction with an unknown environment, a world model is progressively constructed using gradient descent. For deriving optimal actions with respect to future reinforcement, planning is applied in two steps: an experience network proposes a plan which is subsequently optimized by gradient descent with a chain of world models, so that an optimal reinforcement may be obtained when it is actually run. The appropriateness of this method is demonstrated by a robotics application and a pole balancing task.

• Type: Conference
• Year: 1990
• Where: NIPS
• Authors: Sebastian Thrun, Knut Möller, Alexander Linden
• Keywords: Connectionist Planning Method, Gradient Descent, NIPS 1990, NIPS 2007, World Models