Random graph null models have found widespread application in diverse research communities analyzing network datasets, including social, information, and economic networks, as well food webs, protein-protein interactions, neuronal networks. The most popular random models, called configuration are defined uniform distributions over a space of graphs with fixed degree sequence. Commonly, properties an empirical compared to ensemble from model order quantify whether meaningful or they instead...

Partitioning large graphs is difficult, especially when performed in the limited models of computation afforded to modern scale computing systems. In this work we introduce restreaming graph partitioning and develop algorithms that similarly streaming yet empirically perform as well fully offline algorithms. partitioning, are partitioned serially a single pass. Restreaming motivated by scenarios where approximately same dataset routinely streamed, making it possible transform into an...

We show that for pulse-coupled oscillators a class of phase response curves with both excitation and inhibition exhibit robust convergence to synchrony on arbitrary aperiodic connected graphs delays. describe the basins give explicit bounds times. These results provide new more methods synchronization sensor nets also have biological implications.

We show that a large class of pulse-coupled oscillators converge with high probability from random initial conditions on graphs time delays. Our analysis combines previous local convergence results, probabilistic network analysis, and classification scheme for type-II phase response curves to produce rigorous lower bounds probabilities based density. These results suggest methods the oscillators, provide insights into balance excitation inhibition in operation biological also design...

Random graph null models have found widespread application in diverse research communities analyzing network datasets, including social, information, and economic networks, as well food webs, protein-protein interactions, neuronal networks. The most popular family of random models, called configuration are defined uniform distributions over a space graphs with fixed degree sequence. Commonly, properties an empirical compared to ensemble from model order quantify whether meaningful or they...

We introduce a system of pulse coupled oscillators that can change both their phases and frequencies; prove when there is separation time scales between phase frequency adjustment the converges to exact synchrony on strongly connected graphs with delays. The analysis involves decomposing network into forest tree-like structures capture causality. Furthermore, we provide lower bound for size basin attraction immediate implications empirical networks random graph models. These results robust...

`Double edge swaps' transform one graph into another while preserving the graph's degree sequence, and have thus been used in a number of popular Markov chain Monte Carlo (MCMC) sampling techniques. However, double edge-swaps can transform, for any fixed two graphs inside classes simple graphs, multigraphs, pseudographs, this is not true which allow self-loops but multiedges (loopy graphs). Indeed, we exactly characterize sequences where swaps cannot reach every valid loopy develop an...

With sufficient time, double edge-swap Markov chain Monte Carlo (MCMC) methods are able to sample uniformly at random from many different and important graph spaces. For instance, for a fixed degree sequence, MCMC can any from: simple graphs; multigraphs (which may have multiedges); pseudographs multiedges and/or multiple self-loops). In this note we extend these `multiloop-graphs', which allow self-loops but not `loopy-multigraphs' single self-loops. We demonstrate that there sequences on...

Algorithmic education theory examines, among other things, the trade-off between reviewing old material and studying new material: time spent learning comes at expense of solidifying one's understanding old. This is captured in `Slow Flashcard System' (SFS) -- a system that has been studied not only for its applications educational software but also critical properties; it simple discrete deterministic capable remarkable complexity, with standing conjectures regarding longterm behavior. Here...

Altruism is typically associated with traits or behaviors that benefit the population as a whole, but are costly to individual. We propose that, when environment rapidly changing, senescence (age-related deterioration) can be altruistic. According numerical simulations of an agent-based model, while long-lived individuals outcompete their short lived peers, populations composed more likely go extinct during periods rapid environmental change. Moreover, in many situations where other...

Abstract Many psychiatric disorders are marked by impaired decision-making during an approach-avoidance conflict. Current experiments elicit approachavoidance conflicts in bandit tasks pairing individual’s actions with consequences that simultaneously desirable (reward) and undesirable (harm). We frame conflict as a multi-objective multi-armed bandit. By defining general decision-maker limiting sequence of actions, we disentangle the decision process from learning. Each maker can then be...

A study of hidden risks, anxieties and barriers to STEM student transfer from community college a large, comprehensive university is presented. This qualitative employed thematic analysis responses semi-structured interview that asked students about their hesitancy 4-year institution. Participants included enrolled in collaborative NSF-sponsored S-STEM scholarship project between three colleges large public (all the same metropolitan area). The elements are well documented literature...

Recent experiments and theories of human decision-making suggest positive negative errors are processed encoded differently by serotonin dopamine, with possibly serving to oppose dopamine protect against risky decisions. We introduce a temporal difference (TD) model account for these features. Our involves two critics, an optimistic learning system pessimistic system, whose predictions integrated in time control how potential decisions compete be selected. predicts that can decomposed along...

Abstract Recent experiments and theories of human decision-making suggest positive negative errors are processed encoded differently by serotonin dopamine, with possibly serving to oppose dopamine protect against risky decisions. We introduce a temporal difference (TD) model account for these features. Our involves two critics, an optimistic learning system pessimistic system, whose predictions integrated in time control how potential decisions compete be selected. predicts that can...

Abstract Previous network models have imagined that connections change to promote structural balance, or reflect hierarchies. We propose a model where agents adjust their appear credible an external judge observer. In particular, we envision signed, directed positive edges represent endorsements trust and negative accusations doubt, consider both the strategies might use identify nodes then by changing outgoing edges. First, establish may be able exactly set of ‘honest’ from adversarial...

Increasing opportunities for access to traditional financial systems.

We discuss several interesting random network models which exhibit (provable) explosive transitions and their applications.