Giuseppe Rizzo

Papers

Total Citations

H-Index

About

Giuseppe Rizzo is a researcher at the forefront of making artificial intelligence more interpretable and robust, with a primary focus on natural language understanding for robotic systems. His work addresses a critical challenge in modern AI: the "black box" nature of deep neural networks. Rizzo’s key contribution lies in mitigating algorithmic bias by integrating structured knowledge bases into deep learning pipelines. In his most cited work, he demonstrates how to enhance an LSTM-based semantic parser by grounding it in FrameNet, a rich lexical database of semantic frames. This approach not only improves the parser’s accuracy for spoken language understanding in robots but also ensures the model learns correct, unbiased patterns rather than spurious correlations. By bridging symbolic knowledge representation with neural learning, Rizzo’s research offers a practical pathway toward more trustworthy and explainable AI systems. His work is particularly relevant for human-robot interaction, where reliable language comprehension is paramount. Though still early in his career, Rizzo’s innovative fusion of knowledge bases and deep nets is paving the way for a new generation of AI that is both powerful and principled.