
In this paper, we study Monte Carlo estimators based on the likelihood ratio
approach for steadystate sensitivity. We first extend the result of Glynn and
OlveraCravioto [arXiv:1707.02659] to the setting of continuous time Markov
chains (CTMC) with a countable state space which include models such as
stochastic reaction kinetics and kinetic Monte Carlo lattice system. Then we
show that the variance of the centered LR estimators do not grow in time. This
result suggests that the centered estimators should be favored when the mixing
time of the CTMC is large. We demonstrate a practical implication of this
analysis on a numerical benchmark of two examples for the biochemical reaction
networks.

Holes confined in quantum dots have gained considerable interest in the past
few years due to their potential as spin qubits. Here we demonstrate double
quantum dot devices in Ge hut wires. Low temperature transport measurements
reveal Pauli spin blockade. We demonstrate electricdipole spin resonance by
applying a radio frequency electric field to one of the electrodes defining the
double quantum dot. Next, we induce coherent hole spin oscillations by varying
the duration of the microwave burst. Rabi oscillations with frequencies
reaching 140MHz are observed. Finally, Ramsey experiments reveal dephasing
times of 130ns. The reported results emphasize the potential of Ge as a
platform for fast and scalable hole spin qubit devices.

Privacypreserving releasing of complex data (e.g., image, text, audio)
represents a longstanding challenge for the data mining research community.
Due to rich semantics of the data and lack of a priori knowledge about the
analysis task, excessive sanitization is often necessary to ensure privacy,
leading to significant loss of the data utility. In this paper, we present
dpGAN, a general private releasing framework for semanticrich data. Instead
of sanitizing and then releasing the data, the data curator publishes a deep
generative model which is trained using the original data in a differentially
private manner; with the generative model, the analyst is able to produce an
unlimited amount of synthetic data for arbitrary analysis tasks. In contrast of
alternative solutions, dpGAN highlights a set of key features: (i) it provides
theoretical privacy guarantee via enforcing the differential privacy principle;
(ii) it retains desirable utility in the released model, enabling a variety of
otherwise impossible analyses; and (iii) most importantly, it achieves
practical training scalability and stability by employing multifold
optimization strategies. Through extensive empirical evaluation on benchmark
datasets and analyses, we validate the efficacy of dpGAN.

While robust standard errors and related facilities are available in R for
many types of statistical models, the facilities are notably lacking for models
estimated via lme4. This is because the necessary statistical output, including
the Hessian and casewise gradient of random effect parameters, is not
immediately available from lme4 and is not trivial to obtain. In this article,
we supply and describe two new functions to obtain this output from Gaussian
mixed models: estfun.lmerMod() and vcov.full.lmerMod(). We discuss the
theoretical results implemented in the code, focusing on calculation of robust
standard errors via package sandwich. We also use the Sleepstudy data to
illustrate the code and compare it to a benchmark from package lavaan.

Network monitoring and measurement are crucial in network management to
facilitate quality of service routing and performance evaluation. Software
Defined Networking (SDN) makes network management easier by separating the
control plane and data plane. Network monitoring in SDN is lightweight as
operators only need to install a monitoring module into the controller. Active
monitoring techniques usually introduce too many overheads into the network.
The stateoftheart approaches utilize sampling method, aggregation flow
statistics and passive measurement techniques to reduce overheads. However,
little work in literature has focus on reducing the communication cost of
network monitoring. Moreover, most of the existing approaches select the
polling switch nodes by suboptimal local heuristics. Inspired by the
visibility and central control of SDN, we propose FlowCover, a lowcost
highaccuracy monitoring scheme to support various network management tasks. We
leverage the global view of the network topology and active flows to minimize
the communication cost by formulating the problem as a weighted set cover,
which is proved to be NPhard. Heuristics are presented to obtain the polling
scheme efficiently and handle flow changes practically. We build a simulator to
evaluate the performance of FlowCover. Extensive experiment results show that
FlowCover reduces roughly 50% communication cost without loss of accuracy in
most cases.

Network monitoring and measurement are crucial in network management to
facilitate quality of service routing and performance evaluation. Software
Defined Networking (SDN) makes network management easier by separating the
control plane and data plane. Network monitoring in SDN is relatively
lightweight since operators only need to install a monitoring module into the
controller. Active monitoring techniques usually introduce extra overhead into
the network. The stateoftheart approaches utilize sampling, aggregation and
passive measurement techniques to reduce measurement overhead. However, little
work has focused on reducing the communication cost of network monitoring.
Moreover, most of the existing approaches select polling switch nodes by
suboptimal local heuristics.Inspired by the visibility and central control of
SDN, we propose CeMon, a generic lowcost highaccuracy monitoring system that
supports various network management tasks. We first propose a Maximum Coverage
Polling Scheme (MCPS) to optimize the polling cost for all active flows. The
problem is formulated as a weighted set cover problem which is proved to be
NPhard. Heuristics are presented to obtain the polling scheme efficiently and
handle traffic dynamics practically. In order to balance the cost and
flexibility, an Adaptive Finegrained Polling Scheme (AFPS) is proposed as a
complementary method to implement flow level measurement tasks. Three sampling
algorithms are further proposed to eliminate measurement overhead while
maintain high accuracy. Both emulation and simulation results show that MCPS
reduces more than 50% of the communication cost with negligible loss of
accuracy for different topologies and traffics. We also use real traces to
demonstrate that AFPS reduces the cost by up to 20% with only 1.5% loss in
accuracy.

Many of today's machine learning (ML) systems are not built from scratch, but
are compositions of an array of {\em modular learning components} (MLCs). The
increasing use of MLCs significantly simplifies the ML system development
cycles. However, as most MLCs are contributed and maintained by third parties,
their lack of standardization and regulation entails profound security
implications.
In this paper, for the first time, we demonstrate that potentially harmful
MLCs pose immense threats to the security of ML systems. We present a broad
class of {\em logicbomb} attacks in which maliciously crafted MLCs trigger
host systems to malfunction in a predictable manner. By empirically studying
two stateoftheart ML systems in the healthcare domain, we explore the
feasibility of such attacks. For example, we show that, without prior knowledge
about the host ML system, by modifying only 3.3{\textperthousand} of the MLC's
parameters, each with distortion below $10^{3}$, the adversary is able to
force the misdiagnosis of target victims' skin cancers with 100\% success rate.
We provide analytical justification for the success of such attacks, which
points to the fundamental characteristics of today's ML models: high
dimensionality, nonlinearity, and nonconvexity. The issue thus seems
fundamental to many ML systems. We further discuss potential countermeasures to
mitigate MLCbased attacks and their potential technical challenges.

Stochastic reaction networks that exhibit bistability are common in many
fields such as systems biology and materials science. Sampling of the
stationary distribution is crucial for understanding and characterizing the
long term dynamics of bistable stochastic dynamical systems. However, this is
normally hindered by the insufficient sampling of the rare transitions between
the two metastable regions. In this paper, we apply the parallel replica
(ParRep) method for continuous time Markov chain to accelerate the stationary
distribution sampling of bistable stochastic reaction networks. The proposed
method uses parallel computing to accelerate the sampling of rare transitions
and it is very easy to implement. We combine ParRep with the path space
information bounds for parametric sensitivity analysis. We demonstrate the
efficiency and accuracy of the method by studying the Schl\"{o}gl model and the
genetic switches network.

With the increasing demands of applications in virtual reality such as 3D
films, virtual HumanMachine Interactions and virtual agents, the analysis of
3D human face analysis is considered to be more and more important as a
fundamental step for those virtual reality tasks. Due to information provided
by an additional dimension, 3D facial reconstruction enables aforementioned
tasks to be achieved with higher accuracy than those based on 2D facial
analysis. The denser the 3D facial model is, the more information it could
provide. However, most existing dense 3D facial reconstruction methods require
complicated processing and high system cost. To this end, this paper presents a
novel method that simplifies the process of dense 3D facial reconstruction by
employing only one frame of depth data obtained with an offtheshelf RGBD
sensor. The experiments showed competitive results with real world data.

Motivation: Single cell transcriptome sequencing (scRNASeq) has become a
revolutionary tool to study cellular and molecular processes at single cell
resolution. Among existing technologies, the recently developed dropletbased
platform enables efficient parallel processing of thousands of single cells
with direct counting of transcript copies using Unique Molecular Identifier
(UMI). Despite the technology advances, statistical methods and computational
tools are still lacking for analyzing dropletbased scRNASeq data.
Particularly, modelbased approaches for clustering largescale single cell
transcriptomic data are still underexplored. Methods: We developed DIMMSC, a
Dirichlet Mixture Model for clustering dropletbased Single Cell transcriptomic
data. This approach explicitly models UMI count data from scRNASeq experiments
and characterizes variations across different cell clusters via a Dirichlet
mixture prior. An expectationmaximization algorithm is used for parameter
inference. Results: We performed comprehensive simulations to evaluate DIMMSC
and compared it with existing clustering methods such as Kmeans, CellTree and
Seurat. In addition, we analyzed public scRNASeq datasets with known cluster
labels and inhouse scRNASeq datasets from a study of systemic sclerosis with
prior biological knowledge to benchmark and validate DIMMSC. Both simulation
studies and real data applications demonstrated that overall, DIMMSC achieves
substantially improved clustering accuracy and much lower clustering
variability compared to other existing clustering methods. More importantly, as
a modelbased approach, DIMMSC is able to quantify the clustering uncertainty
for each single cell, facilitating rigorous statistical inference and
biological interpretations, which are typically unavailable from existing
clustering methods.

We propose two algorithms for simulating continuous time Markov chains in the
presence of metastability. We show that the algorithms correctly estimate,
under the ergodicity assumption, stationary averages of the process. Both
algorithms, based on the idea of the parallel replica method, use parallel
computing in order to explore metastable sets more efficiently. The algorithms
require no assumptions on the Markov chains beyond ergodicity and the presence
of identifiable metastability.
In particular, there is no assumption on reversibility. We present error
analyses, as well as numerical simulations on multiscale stochastic reaction
network models in order to demonstrate consistency of the method and its
efficiency.

Human creativity is the ultimate driving force behind scientific progress.
While the building blocks of innovations are often embodied in existing
knowledge, it is creativity that blends seemingly disparate ideas. Existing
studies have made striding advances in quantifying creativity of scientific
publications by investigating their citation relationships. Yet, little is
known hitherto about the underlying mechanisms governing scientific creative
processes, largely due to that a paper's references, at best, only partially
reflect its authors' actual information consumption. This work represents an
initial step towards finegrained understanding of creative processes in
scientific enterprise. In specific, using two webscale longitudinal datasets
(120.1 million papers and 53.5 billion web requests spanning 4 years), we
directly contrast authors' information consumption behaviors against their
knowledge products. We find that, of 59.0\% papers across all scientific
fields, 25.7\% of their creativity can be readily explained by information
consumed by their authors. Further, by leveraging these findings, we develop a
predictive framework that accurately identifies the most critical knowledge to
fostering target scientific innovations. We believe that our framework is of
fundamental importance to the study of scientific creativity. It promotes
strategies to stimulate and potentially automate creative processes, and
provides insights towards more effective designs of information recommendation
platforms.

The Massive Open Online Course (MOOC) has expanded significantly in recent
years. With the widespread of MOOC, the opportunity to study the fascinating
courses for free has attracted numerous people of diverse educational
backgrounds all over the world. In the big data era, a key research topic for
MOOC is how to mine the needed courses in the massive course databases in cloud
for each individual student accurately and rapidly as the number of courses is
increasing fleetly. In this respect, the key challenge is how to realize
personalized course recommendation as well as to reduce the computing and
storage costs for the tremendous course data. In this paper, we propose a big
datasupported, contextaware online learningbased course recommender system
that could handle the dynamic and infinitely massive datasets, which recommends
courses by using personalized context information and historical statistics.
The contextawareness takes the personal preferences into consideration, making
the recommendation suitable for people with different backgrounds. Besides, the
algorithm achieves the sublinear regret performance, which means it can
gradually recommend the mostly preferred and matched courses to students. In
addition, our storage module is expanded to the distributedconnected storage
nodes, where the devised algorithm can handle massive course storage problems
from heterogeneous sources of course datasets. Comparing to existing
algorithms, our proposed algorithms achieve the linear time complexity and
space complexity. Experiment results verify the superiority of our algorithms
when comparing with existing ones in the MOOC big data setting.

Most common Monte Carlo methods for sensitivity analysis of stochastic
reaction networks are the finite difference (FD), the Girsanov transformation
(GT) and the regularized pathwise derivative (RPD) methods. It has been
numerically observed in the literature, that the biased FD and RPD methods tend
to have lower variance than the unbiased GT method and that centering the GT
method (CGT) reduces its variance. We provide a theoretical justification for
these observations in terms of system size asymptotic analysis under what is
known as the classical scaling. Our analysis applies to GT, CGT and FD, and
shows that the standard deviations of their estimators when normalized by the
actual sensitivity, scale as $\mathcal{O}(N^{1/2}), \mathcal{O}(1)$ and
$\mathcal{O}(N^{1/2})$ respectively, as system size $N \to \infty$. In the
case of the FD methods, the $N \to \infty$ asymptotics are obtained keeping the
finite difference perturbation $h$ fixed. Our numerical examples verify that
our order estimates are sharp and that the variance of the RPD method scales
similarly to the FD methods. We combine our large $N$ asymptotics with
previously known small $h$ asymptotics to obtain the best choice of $h$ in
terms of $N$, and estimate the number $N_s$ of simulations required to achieve
a prescribed relative $\mathcal{L}_2$ error $\delta$. This shows that $N_s$
depends on $\delta$ and $N$ as $\delta^{2  \frac{\gamma_2}{\gamma_1}} N^{1},
\delta^{2}$ and $N \delta^{2}$, for FD, CGT and GT respectively. Here
$\gamma_1 >0, \gamma_2>0$ depend on the type of FD method used.

Some rigorous results and statistics of the solution space of VertexCovers
on bipartite graphs are given in this paper. Based on the $K\ddot{o}nig$'s
theorem, an exact solution space expression algorithm is proposed and
statistical analysis of the nodes' states is provided. The statistical results
fit well with the algorithmic results until the emergence of the unfrozen core,
which makes the fluctuation of statistical quantities and causes the replica
symmetric breaking in the solutions. Besides, the entropy of bipartite
VertexCover solutions is calculated with the clustering entropy using a cycle
simplification technique for the unfrozen core. Furthermore, as generalization
of bipartite graphs, bipartite core graph is proposed, the solution space of
which can also be easily determined; and based on these results, how to
generate a $K\ddot{o}nigEgerv\acute{a}ry$ subgraph is studied by a growth
process of adding edges. The investigation of solution space of bipartite graph
VertexCover provides intensive understanding and some insights on the solution
space complexity, and will produce benefit for finding maximal
$K\ddot{o}nigEgerv\acute{a}ry$ subgraphs, solving general graph VertexCover
and recognizing the intrinsic hardness of NPcomplete problems.

When pulling a probe particle in a manyparticle system with fixed velocity,
the probe's effective friction, defined as average pulling force over its
velocity, $\gamma_{eff}:=\langle F_{ex}\rangle/u$, first keeps constant (linear
response), then decreases (thinning) and finally increases (thickening). We
propose a threetimescales picture (TTSP) to unify thinning and thickening
behaviour. The points of the TTSP are that there are three distinct time scales
of bath particles: diffusion, damping, and single probebath (PB) collision;
the dominating time scales, which are controlled by the pulling velocity,
determine the behaviour of the probe's friction. We confirm the TTSP by
Langevin dynamics simulation. Microscopically, we find that for computing the
effective friction, Maxwellian distribution of bath particles' velocities works
in low Reynolds number (Re) but fails in high Re. It can be understood based on
the microscopic mechanism of thickening obtained in the $T=0$ limit. Based on
the TTSP, we explain different thinning and thickening observations in some
earlier literature.

This paper is based on our previous work on neural coding. It is a
selforganized model supported by existing evidences. Firstly, we briefly
introduce this model in this paper, and then we explain the neural mechanism of
language and reasoning with it. Moreover, we find that the position of an area
determines its importance. Specifically, language relevant areas are in the
capital position of the cortical kingdom. Therefore they are closely related
with autonomous consciousness and working memories. In essence, language is a
miniature of the real world. Briefly, this paper would like to bridge the gap
between molecule mechanism of neurons and advanced functions such as language
and reasoning.

Proteinprotein interaction extraction is the key precondition of the
construction of protein knowledge network, and it is very important for the
research in the biomedicine. This paper extracted directional proteinprotein
interaction from the biological text, using the SVMbased method. Experiments
were evaluated on the LLL05 corpus with good results. The results show that
dependency features are import for the proteinprotein interaction extraction
and features related to the interaction word are effective for the interaction
direction judgment. At last, we analyzed the effects of different features and
planed for the next step.

Based on existing data, we wish to put forward a biological model of motor
system on the neuron scale. Then we indicate its implications in statistics and
learning. Specifically, neuron firing frequency and synaptic strength are
probability estimates in essence. And the lateral inhibition also has
statistical implications. From the standpoint of learning, dendritic
competition through retrograde messengers is the foundation of conditional
reflex and grandmother cell coding. And they are the kernel mechanisms of motor
learning and sensory motor integration respectively. Finally, we compare motor
system with sensory system. In short, we would like to bridge the gap between
molecule evidences and computational models.

The coding mechanism of sensory memory on the neuron scale is one of the most
important questions in neuroscience. We have put forward a quantitative neural
network model, which is self organized, self similar, and self adaptive, just
like an ecosystem following Darwin theory. According to this model, neural
coding is a mult to one mapping from objects to neurons. And the whole cerebrum
is a realtime statistical Turing Machine, with powerful representing and
learning ability. This model can reconcile some important disputations, such
as: temporal coding versus rate based coding, grandmother cell versus
population coding, and decay theory versus interference theory. And it has also
provided explanations for some key questions such as memory consolidation,
episodic memory, consciousness, and sentiment. Philosophical significance is
indicated at last.

We have put forwards a unified quantitative framework of vision and audition,
based on existing data and theories. According to this model, the retina is a
feedforward network selfadaptive to inputs in a specific period. After fully
grown, cells become specialized detectors based on statistics of stimulus
history. This model has provided explanations for perception mechanisms of
colour, shape, depth and motion. Moreover, based on this ground we have put
forwards a bold conjecture that single ear can detect sound direction. This is
complementary to existing theories and has provided better explanations for
sound localization.

When pulling a particle in a driven granular fluid with constant force
$F_{ex}$, the probe particle approaches a steadystate average velocity $v$.
This velocity and the corresponding friction coefficient of the probe
$\zeta=F_{ex}/v$ are obtained within a schematic model of modecoupling theory
and compared to results from eventdriven simulations. For small and moderate
drag forces, the model describes the simulation results successfully for both
the linear as well as the nonlinear region: The linear response regime
(constant friction) for small drag forces is followed by shear thinning
(decreasing friction) for moderate forces. For large forces, the model
demonstrates a subsequent increasing friction in qualitative agreement with the
data. The squareroot increase of the friction with force found in [Fiege et
al., Granular Matter $\boldsymbol{14}$, 247 (2012)] is explained by a simple
kinetic theory.

In recent years, a market for mortality derivatives began developing as a way
to handle systematic mortality risk, which is inherent in life insurance and
annuity contracts. Systematic mortality risk is due to the uncertain
development of future mortality intensities, or {\it hazard rates}. In this
paper, we develop a theory for pricing pure endowments when hedging with a
mortality forward is allowed. The hazard rate associated with the pure
endowment and the reference hazard rate for the mortality forward are
correlated and are modeled by diffusion processes. We price the pure endowment
by assuming that the issuing company hedges its contract with the mortality
forward and requires compensation for the unhedgeable part of the mortality
risk in the form of a prespecified instantaneous Sharpe ratio. The major
result of this paper is that the value per contract solves a linear partial
differential equation as the number of contracts approaches infinity. One can
represent the limiting price as an expectation under an equivalent martingale
measure. Another important result is that hedging with the mortality forward
may raise or lower the price of this pure endowment comparing to its price
without hedging, as determined in Bayraktar et al. [2009]. The market price of
the reference mortality risk and the correlation between the two portfolios
jointly determine the cost of hedging. We demonstrate our results using
numerical examples.

We find the minimum probability of lifetime ruin of an investor who can
invest in a market with a risky and a riskless asset and who can purchase a
reversible life annuity. The surrender charge of a life annuity is a proportion
of its value. Ruin occurs when the total of the value of the risky and riskless
assets and the surrender value of the life annuity reaches zero. We find the
optimal investment strategy and optimal annuity purchase and surrender
strategies in two situations: (i) the value of the risky and riskless assets is
allowed to be negative, with the imputed surrender value of the life annuity
keeping the total positive; or (ii) the value of the risky and riskless assets
is required to be nonnegative. In the first case, although the individual has
the flexiblity to buy or sell at any time, we find that the individual will not
buy a life annuity unless she can cover all her consumption via the annuity and
she will never sell her annuity. In the second case, the individual surrenders
just enough annuity income to keep her total assets positive. However, in this
second case, the individual's annuity purchasing strategy depends on the size
of the proportional surrender charge. When the charge is large enough, the
individual will not buy a life annuity unless she can cover all her
consumption, the socalled safe level. When the charge is small enough, the
individual will buy a life annuity at a wealth lower than this safe level.

We study pairwise entanglements in spinhalf and spinone Heisenberg chains
with an open boundary condition, respectively. We find out that the
groundstate and the firstexcitedstate entanglements are equal for the
threesite spinone chain. When the number of sites L>3, the concurrences and
negativities display oscillatory behaviors, and the oscillations of the
groundstate and the firstexcitedstate entanglements are out of phase or in
phase.