Learning programmatically structured representations with perceptor gradients
S Penkov, S Ramamoorthy - arXiv preprint arXiv:1905.00956, 2019 - arxiv.org
arXiv preprint arXiv:1905.00956, 2019•arxiv.org
We present the perceptor gradients algorithm--a novel approach to learning symbolic
representations based on the idea of decomposing an agent's policy into i) a perceptor
network extracting symbols from raw observation data and ii) a task encoding program
which maps the input symbols to output actions. We show that the proposed algorithm is
able to learn representations that can be directly fed into a Linear-Quadratic Regulator
(LQR) or a general purpose A* planner. Our experimental results confirm that the perceptor …
representations based on the idea of decomposing an agent's policy into i) a perceptor
network extracting symbols from raw observation data and ii) a task encoding program
which maps the input symbols to output actions. We show that the proposed algorithm is
able to learn representations that can be directly fed into a Linear-Quadratic Regulator
(LQR) or a general purpose A* planner. Our experimental results confirm that the perceptor …
We present the perceptor gradients algorithm -- a novel approach to learning symbolic representations based on the idea of decomposing an agent's policy into i) a perceptor network extracting symbols from raw observation data and ii) a task encoding program which maps the input symbols to output actions. We show that the proposed algorithm is able to learn representations that can be directly fed into a Linear-Quadratic Regulator (LQR) or a general purpose A* planner. Our experimental results confirm that the perceptor gradients algorithm is able to efficiently learn transferable symbolic representations as well as generate new observations according to a semantically meaningful specification.
arxiv.org
Showing the best result for this search. See all results