State Space LSTM Models with Particle MCMC Inference
This work addresses the need for more interpretable sequence models in machine learning, though it appears incremental as it builds on prior research combining topic models with LSTMs.
The paper tackles the problem of combining the interpretability of state space models with the performance of LSTMs by introducing State Space LSTM models, and it shows that an efficient sequential Monte Carlo inference algorithm achieves superior and stable results across various domains.
Long Short-Term Memory (LSTM) is one of the most powerful sequence models. Despite the strong performance, however, it lacks the nice interpretability as in state space models. In this paper, we present a way to combine the best of both worlds by introducing State Space LSTM (SSL) models that generalizes the earlier work \cite{zaheer2017latent} of combining topic models with LSTM. However, unlike \cite{zaheer2017latent}, we do not make any factorization assumptions in our inference algorithm. We present an efficient sampler based on sequential Monte Carlo (SMC) method that draws from the joint posterior directly. Experimental results confirms the superiority and stability of this SMC inference algorithm on a variety of domains.