QNLP in Practice: Running Compositional Models of Meaning on a Quantum Computer

Quantum Natural Language Processing (QNLP) deals with the design and implementation of NLP models intended to be run on quantum hardware. In this paper, we present results first experiments conducted Noisy Intermediate-Scale (NISQ) computers for datasets size greater than 100 sentences. Exploiting formal similarity compositional model meaning by Coecke, Sadrzadeh Clark (2010) theory, create representations sentences that have a natural mapping circuits. We use these implement successfully train solve simple sentence classification tasks conduct simulations compare syntax-sensitive Coecke et al. two baselines less or no syntax; specifically, analogues "bag-of-words" model, where syntax is not taken into account at all, word-sequence only word order respected. demonstrate all converge smoothly both in when hardware, are expected ones based nature used. Another important goal paper describe way accessible AI researchers main principles, process challenges Our aim doing take small steps unexplored research territory pave practical Processing.