Deep learning in nano-photonics: inverse design and beyond

Modeling of nano-photonics

The ability of photonic crystals to mold the flow of light in new ways can lead to a variety of novel and improved designs of optical nanocomponents and nanodevices in photonics. Two examples will be presented: a) Using linear materials, a polarization independent waveguide is designed in a 3D photonic crystal. It is demonstrated that this system provides lossless guiding of light at length-sca...

Beyond Structured Prediction: Inverse Reinforcement Learning

Inverse Reinforcement Learning via Deep Gaussian Process

We propose a new approach to inverse reinforcement learning (IRL) based on the deep Gaussian process (deep GP) model, which is capable of learning complicated reward structures with few demonstrations. Our model stacks multiple latent GP layers to learn abstract representations of the state feature space, which is linked to the demonstrations through the Maximum Entropy learning framework. Inco...

Maximum Entropy Deep Inverse Reinforcement Learning

This paper presents a general framework for exploiting the representational capacity of neural networks to approximate complex, nonlinear reward functions in the context of solving the inverse reinforcement learning (IRL) problem. We show in this context that the Maximum Entropy paradigm for IRL lends itself naturally to the efficient training of deep architectures. At test time, the approach l...

Deep Convolutional Framelets: A General Deep Learning for Inverse Problems

Recently, deep learning approaches with various network architectures have achieved significant performance improvement over existing iterative reconstruction methods in various imaging problems. However, it is still unclear why these deep learning architectures work for specific inverse problems. Moreover, in contrast to the usual evolution of signal processing theory around the classical theo...