Matt Oberdorfer

Learn More
Lesions were produced in the nasal-superior quadrant of rat retinas at 1 day postnatal. Both the optic fiber and ganglion cell layers were destroyed at the lesion site. Retrograde changes in the more peripherally located ganglion cell bodies, their optic fibers, and neuroglia were monitored by light and electron microscopy. No optic fibers remained in the(More)
We present the first reinforcement-learning model to self-improve its reward-modulated training implemented through a continuously improving " intuition " neural network. An agent was trained how to play the arcade video game Pong with two reward-based alternatives, one where the paddle was placed randomly during training, and a second where the paddle was(More)
  • 1