Coupling Matrix Manifolds and Their Applications in Optimal Transport. (arXiv:1911.06905v1 [cs.LG])

Optimal transport (OT) is a powerful tool for measuring the distance between two defined probability distributions. In this paper, we develop a new manifold named the coupling matrix manifold (CMM), where each point on CMM can be regarded as the transportation plan of the OT problem. We firstly explore the Riemannian geometry of CMM with…

How data, synapses and neurons interact with each other: a variational principle marrying gradient ascent and message passing. (arXiv:1911.07662v1 [stat.ML])

Unsupervised learning requiring only raw data is not only a fundamental function of the cerebral cortex, but also a foundation for a next generation of artificial neural networks. However, a unified theoretical framework to treat sensory inputs, synapses and neural activity together is still lacking. The computational obstacle originates from the discrete nature of synapses,…

Comparison of screening for methicillin-resistant Staphylococcus aureus (MRSA) at hospital admission and discharge. (arXiv:1911.07711v1 [q-bio.PE])

Methicillin-resistant Staphylococcus aureus (MRSA) is a significant contributor to the growing concern of antibiotic resistant bacteria, especially given its stubborn persistence in hospitals and other health care facility settings. In combination with this characteristic of S. aureus (colloquially referred to as staph), MRSA presents an additional barrier to treatment and is now believed to have…

Granular Motor State Monitoring of Free Living Parkinson’s Disease Patients via Deep Learning. (arXiv:1911.06913v1 [stat.AP])

Parkinson’s disease (PD) is the second most common neurodegenerative disease worldwide and affects around 1% of the (60+ years old) elderly population in industrial nations. More than 80% of PD patients suffer from motor symptoms, which could be well addressed if a personalized medication schedule and dosage could be administered to them. However, such personalized…

Benanza: Automatic uBenchmark Generation to Compute “Lower-bound” Latency and Inform Optimizations of Deep Learning Models on GPUs. (arXiv:1911.06922v1 [cs.LG])

As Deep Learning (DL) models have been increasingly used in latency-sensitive applications, there has been a growing interest in improving their response time. An important venue for such improvement is to profile the execution of these models and characterize their performance to identify possible optimization opportunities. However, the current profiling tools lack the highly desired…

Generalized Maximum Causal Entropy for Inverse Reinforcement Learning. (arXiv:1911.06928v1 [cs.LG])

We consider the problem of learning from demonstrated trajectories with inverse reinforcement learning (IRL). Motivated by a limitation of the classical maximum entropy model in capturing the structure of the network of states, we propose an IRL model based on a generalized version of the causal entropy maximization problem, which allows us to generate a…

Semiparametric Estimation of Correlated Random Coefficient Models without Instrumental Variables. (arXiv:1911.06857v1 [econ.EM])

We study a linear random coefficient model where slope parameters may be correlated with some continuous covariates. Such a model specification may occur in empirical research, for instance, when quantifying the effect of a continuous treatment observed at two time periods. We show one can carry identification and estimation without instruments. We propose a semiparametric…

Innovation and Strategic Network Formation. (arXiv:1911.06872v1 [econ.TH])

We study a model of innovation with a large number of firms that create new technologies by combining several discrete ideas. These ideas can be acquired by private investment or via social learning. Firms face a choice between secrecy, which protects existing intellectual property, and openness, which facilitates social learning. These decisions determine interaction rates…

Biophysical characterization of DNA origami nanostructures reveals inaccessibility to intercalation binding sites. (arXiv:1911.07022v1 [physics.bio-ph])

Intercalation of drug molecules into synthetic DNA nanostructures formed through self-assembled origami has been postulated as a valuable future method for targeted drug delivery. This is due to the excellent biocompatibility of synthetic DNA nanostructures, and high potential for flexible programmability including facile drug release into or near to target cells. Such favourable properties may…