Differentiable User Models
This addresses the problem of integrating complex user models into practical, online AI systems for researchers and developers, though it is incremental as it builds on existing likelihood-free inference methods.
The paper tackles the computational inefficiency of advanced cognitive user models in machine learning pipelines by introducing differentiable surrogates, achieving modeling capabilities comparable to existing methods with a computational cost suitable for online applications, such as reducing hours of computation to real-time in a menu-search task.
Probabilistic user modeling is essential for building machine learning systems in the ubiquitous cases with humans in the loop. However, modern advanced user models, often designed as cognitive behavior simulators, are incompatible with modern machine learning pipelines and computationally prohibitive for most practical applications. We address this problem by introducing widely-applicable differentiable surrogates for bypassing this computational bottleneck; the surrogates enable computationally efficient inference with modern cognitive models. We show experimentally that modeling capabilities comparable to the only available solution, existing likelihood-free inference methods, are achievable with a computational cost suitable for online applications. Finally, we demonstrate how AI-assistants can now use cognitive models for online interaction in a menu-search task, which has so far required hours of computation during interaction.