An information-theoretic approach to intentionality proposing that conscious representations must be unambiguous, with experiments on artificial neural networks demonstrating a method for measuring representational ambiguity.

This work puts forward a novel continual learning algorithm based on brain-inspired principles of neural computation. Contrary to many existing approaches, our method does not make use of task boundaries or data replay.

We present Darts, a Python machine learning library for time series, with a focus on forecasting. Darts offers a variety of models, from classics such as ARIMA to state-of-the-art deep neural networks.