What We Do
Algoritica conducts research in the field of Machine Intelligence. Algoritica has several research goals:
1, understand how networks of computational units cooperate to produce sophisticated
and intelligent behaviour; 2, explore intrinsically motivated agency and unsupervised learning;
3, apply modern techniques from Artificial Intelligence and Machine Learning to
problems in various domains, in order to understand limitations of the techniques, advance the state of the art,
and find opportunities for new applications.
Computing with Networks
Artificial Neural Networks represent just one kind of network-based computational model. To date, there has not
been very extensive investigation of the computational properties of networks of processing units in general. We
aim to shed more light on this research question through pursuit of our first research goal.
Intelligence in nature is embodied in agents that interact with their environment. From single cells to various
forms of animal life, intelligent behaviour emerges as a spontaneous, self-organized phenomenon. Where do the
intrinsically motivated goals that characterize biological teleonomy come from? We believe that answering this
question will be key to realizing artificial agency, and is the aim of our second research goal.
Applied Machine Intelligence
Deep Learning, when combined with techniques such as Reinforcement Learning and Monte Carlo Tree Search, represents
a significant advance in the state of the art of Artificial Intelligence. Algoritica aims to apply these
technologies to various domains, in pursuit of our third research goal.
How to Contact Us
Algoritica is based in Toronto, Canada. Please feel free to
contact us through this website.