Sorry, you need to enable JavaScript to visit this website.

This work proposes a model for continual learning on tasks involving temporal sequences, specifically, human motions. It improves on a recently proposed brain-inspired replay model (BI-R) by building a biologically-inspired conditional temporal variational autoencoder (BI-CTVAE), which instantiates a latent mixture-of-Gaussians for class representation. We investigate a novel continual-learning-to-generate (CL2Gen) scenario where the model generates motion sequences of different classes. The generative accuracy of the model is tested over a set of tasks.


A number of machine learning applications involve time series prediction, and in some cases additional information about dynamical constraints on the target time series may be available. For instance, it might be known that the desired quantity cannot change faster than some rate or that the rate is dependent on some known factors. However, incorporating these constraints into deep learning models, such as recurrent neural networks, is not straightforward.


Deep neural networks (DNNs) allow digital receivers to learn operating in complex environments.
DNNs infer reliably when applied in a similar statistical relationship as the one under which it was trained.
This property constitutes a major drawback of using DNN-aided receivers for dynamic communication systems, whose input-output relationship varies over time.
In such setups, DNN-aided receivers may be required to retrain periodically, which conventionally involves excessive pilot signaling at the cost of reduced spectral efficiency.


This work proposes a novel and scalable reinforcement learning approach for routing in ad-hoc wireless networks. In most previous reinforcement learning based routing methods, the links in the network are assumed to be fixed, and a different agent is trained for


Online prediction for streaming time series data has practical use for many real-world applications where downstream decisions depend on accurate forecasts for the future. Deployment in dynamic environments requires models to adapt quickly to changing data distributions without overfitting. We propose POLA (Predicting Online by Learning rate Adaptation) to automatically regulate the learning rate of recurrent neural network models to adapt to changing time series patterns across time.