Deep Learning
Supplementary Material: Reverse engineering recurrent neural networks with Jacobian switching linear dynamical systems
In general, we have found the JSLDS loss function strengths to be relatively easy to select (see example settings in the specific experiment sections below). RNN's fixed points or slow points would defeat the primary purpose of the method. However, other variations are possible. We set the number of timesteps T = 25. We trained both methods with the Adam optimizer with default settings.
Reverse engineering recurrent neural networks with Jacobian switching linear dynamical systems
They learn to transform input sequences into output sequences by using an internal state that allows data from the past to influence the current state.
Learning to Generate Inversion-Resistant Model Explanations
Figure 1 compares the result of MI attacks using model predictions alone (PredMI) and explanation-aware MI attacks using model predictions and explanations together (ExpMI).
Merging Models with Fisher-Weighted Averaging
Averaging the parameters of models that have the same architecture and initialization can provide a means of combining their respective capabilities.
Tikhonov Regularization is Optimal Transport Robust under Martingale Constraints
To realize these novel tools, tractable computational algorithms are proposed.
