This is the GitHub repository of CurioSAI, a research group driven by curiosity in its many forms.

With diverse backgrounds ranging from complex differential geometry to probability, from computer science to biology, we all share a common fascination with AI, sparked by AlphaGo.

Witnessing AlphaGo's inquisitive exploration of new moves in its legendary matches against Lee Sedol, and their effectiveness, convinced us of the immense potential of deep learning and reinforcement learning across all scientific domains.

Driven by the curiosity to understand AlphaGo's inner workings, we developed our first product: SAI, a sensible Artificial Intelligence, a version of AlphaGo capable of optimizing the score, not just the probability of winning.

We are eager to explore all possible applications of deep learning and reinforcement learning. We are interested in all sectors because we believe that curiosity and interdisciplinarity are the foundation of quality scientific research.

We are also convinced that curiosity is one of the fundamental aspects currently missing in AI: when AI becomes curious, we will be significantly closer to AGI.

The name of our group reflects all of this: CurioSAI. In case you're curious about why it's written without a "u", it's the past participle of "curiosare" (to be curious) in Italian.