Timezone: »

Interaction Screening: Efficient and Sample-Optimal Learning of Ising Models
Marc Vuffray · Sidhant Misra · Andrey Lokhov · Michael Chertkov

Mon Dec 05 09:00 AM -- 12:30 PM (PST) @ Area 5+6+7+8 #129 #None

We consider the problem of learning the underlying graph of an unknown Ising model on p spins from a collection of i.i.d. samples generated from the model. We suggest a new estimator that is computationally efficient and requires a number of samples that is near-optimal with respect to previously established information theoretic lower-bound. Our statistical estimator has a physical interpretation in terms of "interaction screening". The estimator is consistent and is efficiently implemented using convex optimization. We prove that with appropriate regularization, the estimator recovers the underlying graph using a number of samples that is logarithmic in the system size p and exponential in the maximum coupling-intensity and maximum node-degree.

Author Information

Marc Vuffray (Los Alamos National Laboratory)

I’m a staff research scientist in the Theoretical Division at the Los Alamos National Laboratory (LANL), New Mexico, where I am part of the Advanced Network Science Initiative (ANSI) as well as the Condensed Matter and Complex Systems Group (T-4). My background is in statistical physics and information theory. My current work focuses on the design of machine learning techniques for learning probabilistic networks and on the development of new methods to control and optimize energy networks under uncertainty.

Sidhant Misra (Los Alamos National Laboratory)
Andrey Lokhov (Los Alamos National Laboratory)
Michael Chertkov (Los Alamos National Laboratory)

More from the Same Authors