Timezone: »

Efficient identification of informative features in simulation-based inference
Jonas Beck · Michael Deistler · Yves Bernaerts · Jakob H Macke · Philipp Berens

Tue Nov 29 09:00 AM -- 11:00 AM (PST) @ Hall J #510

Simulation-based Bayesian inference (SBI) can be used to estimate the parameters of complex mechanistic models given observed model outputs without requiring access to explicit likelihood evaluations. A prime example for the application of SBI in neuroscience involves estimating the parameters governing the response dynamics of Hodgkin-Huxley (HH) models from electrophysiological measurements, by inferring a posterior over the parameters that is consistent with a set of observations. To this end, many SBI methods employ a set of summary statistics or scientifically interpretable features to estimate a surrogate likelihood or posterior. However, currently, there is no way to identify how much each summary statistic or feature contributes to reducing posterior uncertainty. To address this challenge, one could simply compare the posteriors with and without a given feature included in the inference process. However, for large or nested feature sets, this would necessitate repeatedly estimating the posterior, which is computationally expensive or even prohibitive. Here, we provide a more efficient approach based on the SBI method neural likelihood estimation (NLE): We show that one can marginalize the trained surrogate likelihood post-hoc before inferring the posterior to assess the contribution of a feature. We demonstrate the usefulness of our method by identifying the most important features for inferring parameters of an example HH neuron model. Beyond neuroscience, our method is generally applicable to SBI workflows that rely on data features for inference used in other scientific fields.

Author Information

Jonas Beck (University of Tuebingen)
Michael Deistler (University of Tuebingen)
Yves Bernaerts (University of Tübingen)

On the borderline between neural science and machine learning. Using sparse bottleneck neural networks to obtain lower dimensional visualizations of paired datasets. Interested in inferring interpretable biophysical model parameters using deep neural network architectures.

Jakob H Macke (University of Tübingen & MPI IS Tübingen)
Philipp Berens (University of Tübingen)

More from the Same Authors