Thurstonian Boltzmann Machines: Learning from Multiple Inequalities
This provides a unified framework for handling multiple data types in machine learning, though it appears incremental as it builds on existing Boltzmann machine and Thurstonian concepts.
The authors tackled the problem of learning from diverse discrete data types by introducing Thurstonian Boltzmann Machines, which unify them through latent continuous variables and inequalities, achieving versatility in applications like digit recognition, collaborative filtering, and social survey analysis.
We introduce Thurstonian Boltzmann Machines (TBM), a unified architecture that can naturally incorporate a wide range of data inputs at the same time. Our motivation rests in the Thurstonian view that many discrete data types can be considered as being generated from a subset of underlying latent continuous variables, and in the observation that each realisation of a discrete type imposes certain inequalities on those variables. Thus learning and inference in TBM reduce to making sense of a set of inequalities. Our proposed TBM naturally supports the following types: Gaussian, intervals, censored, binary, categorical, muticategorical, ordinal, (in)-complete rank with and without ties. We demonstrate the versatility and capacity of the proposed model on three applications of very different natures; namely handwritten digit recognition, collaborative filtering and complex social survey analysis.