Memory-based parameterization with differentiable solver: Application to Lorenz '96.

Physical parameterizations (or closures) are used as representations of unresolved subgrid processes within weather and global climate models or coarse-scale turbulent models, whose resolutions are too coarse to resolve small-scale processes. These parameterizations are typically grounded on physically based, yet empirical, representations of the underlying small-scale processes. Machine learning-based parameterizations have recently been proposed as an alternative solution and have shown great promise to reduce uncertainties associated with the parameterization of small-scale processes.

Gaussian processes meet NeuralODEs: a Bayesian framework for learning the dynamics of partially observed systems from scarce and noisy data.

We present a machine learning framework (GP-NODE) for Bayesian model discovery from partial, noisy and irregular observations of nonlinear dynamical systems. The proposed method takes advantage of differentiable programming to propagate gradient information through ordinary differential equation solvers and perform Bayesian inference with respect to unknown model parameters using Hamiltonian Monte Carlo sampling and Gaussian Process priors over the observed system states. This allows us to exploit temporal correlations in the observed data, and efficiently infer posterior distributions over plausible models with quantified uncertainty.

Vangl2 in the Dentate Network Modulates Pattern Separation and Pattern Completion.

The organization of spatial information, including pattern completion and pattern separation processes, relies on the hippocampal circuits, yet the molecular and cellular mechanisms underlying these two processes are elusive. Here, we find that loss of Vangl2, a core PCP gene, results in opposite effects on pattern completion and pattern separation processes. Mechanistically, we show that Vangl2 loss maintains young postmitotic granule cells in an immature state, providing increased cellular input for pattern separation.

Deletion of in adult mice impairs basal AMPA receptor transmission and LTP in hippocampal CA1 pyramidal neurons.

Leucine-rich repeat transmembrane (LRRTM) proteins are synaptic cell adhesion molecules that influence synapse formation and function. They are genetically associated with neuropsychiatric disorders, and via their synaptic actions likely regulate the establishment and function of neural circuits in the mammalian brain. Here, we take advantage of the generation of a and double conditional knockout mouse ( cKO) to examine the role of LRRTM1,2 at mature excitatory synapses in hippocampal CA1 pyramidal neurons.