Machine Learning and Artificial Neural Networks
At IDSIA we purse the following main research directions in Machine Learning and Artificial Neural Networks: recurrent neural networks, using machine learning to process temporal data, probabilistic graphical models and causality, reinforcement learning and geometric deep learning.

Recurrent neural networks

The human brain is a recurrent neural network (RNN): a network of neurons with feedback connections. It can learn many behaviors / sequence processing tasks / algorithms / programs that are not learnable by traditional machine learning methods.

Machine learning of temporal data

A time series is a sequence of observations of the same variable, collected over time.
At IDSIA, we cover different research areas related to time series, such as forecasting, time series classification and anomaly detection.

Bayesian Networks (BNs)

Bayesian Networks are foundational model in machine learning and artificial intelligence.

Causal analysis and Knowledge Engineering

To understand the causal relations between model variables, dedicated mathematical tools are required. 

Reinforcement learning and POMDPs

Work at IDSIA has led to the first universal reinforcement learner for essentially arbitrary computable environments. 

Graph and geometric deep learning

Graph and geometric deep learning are machine learning fields that combine graph representations for data and machine learning to exploit the inductive bias associated with the presence of functional dependencies among data 

Security for Machine Learning, Machine Learning for security

 Security is both a key enabler and an extremely relevant application for machine learning