Constraining Galactic Accelerations with Stellar Streams

Metro Toronto Convention Centre 

We present a data-driven method for estimating galactic accelerations from phase-space measurements of stellar streams. Using a differentiable neural network to parameterize the track of the stream in phase-space, our approach enables a direct estimate of the acceleration field in the neighborhood of the stream. A model for the galactic gravitational potential does not need to be specified beforehand. By treating each stream as a collection of proximate orbits, our method utilizes the chain rule to convert derivatives of phase-space coordinates along the stream-track to estimated galactic accelerations. Once acceleration vectors are sampled along the stream, standard analytic models for the Galactic potential can then be constrained. Alternatively, we demonstrate that the potential can be represented with a neural network to enable full model flexibility while minimizing non-physical artifacts through Poisson's equation. On mock data, our approach recovers the true potential with sub-percent level fractional errors across a range of scales, providing a new avenue to map the Milky Way with stellar streams and constrain dark matter on galactic scales.