Inspirational Machine Learning Papers from Other Fields
(a very brief selection of the flood of research in vision/voice/etc)
Both of these works on CLDNNs are an excellent example of how learning from raw time series waveforms, building recurrent time series representations, and then mapping to supervised targets can be done with a CLDNN architecture.
A powerful method for end to end training of localization networks in the image domain by fitting 2D Affine transform parameters through a regression task prior to the discriminative task.
Generative models for acoustic time-series
Google’s WaveNet model for time series acoustic voice synthesis
Radio Machine Learning Papers
Communication System PHY Learning
The fundamental construct of the channel autoencoder for learning to communicate digital information across wireless channels. Learning new physical layer encodings which work as well as traditional expert-defined PHYs but are now continuously adaptive and potentially much lower complexity.
Supervised Learning of modulation types directly on radio time series data using a deep convolutional neural network.
This work looks at extending purely supervised models into the semi-supervised or purely unsupervised areas, since the world generally isn’t labeled nicely for us.
Fundamental Structure Learning
Unsupervised sparse representation learning using a convolutional denoising autoencoder on radio time series data for compression, reconstruction, and compact representation.
Initial work in building radio attention models to provide some degree of synchronization and canonical form signal inputs to the convolutional classification network.
Sequence Model Learning
High level protocol traffic sequence classification based on low level PHY sample data representations through feature learning.
Looking at traditional reconstruction based anomaly detection on time series datasets with Kalman predictors replaced by NN based sequence prediction on wideband RF datasets.
Radio Control System Learning
Learning to optimize radio control and behavior using deep reinforcement learning techniques in Keras. Info about the KeRLym framework for DeepRL using the agent and environment interfaces from OpenAI Gym