arXiv subjects

Lijun Zhang

Publications and source records attributed to Lijun Zhang.

At least 253 records · Page 14Linked to original sources

arXiv2017

Improved Dynamic Regret for Non-degenerate Functions

Recently, there has been a growing research interest in the analysis of dynamic regret, which measures the performance of an online learner against a sequence of local minimizers. By exploiting the strong convexity, previous studies have shown that the dynamic regret can be upper bounded by the path-length of the comparator sequence. In this paper, we illustrate that the dynamic regret can be further improved by allowing the learner to query the gradient of the function multiple times, and meanwhile the strong convexity can be weakened to other non-degenerate conditions. Specifically, we introduce the squared path-length, which could be much smaller than the path-length, as a new regularity of the comparator sequence. When multiple gradients are accessible to the learner, we first demonstrate that the dynamic regret of strongly convex functions can be upper bounded by the minimum of the path-length and the squared path-length. We then extend our theoretical guarantee to functions that are semi-strongly convex or self-concordant. To the best of our knowledge, this is the first time that semi-strong convexity and self-concordance are utilized to tighten the dynamic regret.

cs.LG↗

arXiv2017

A Simple Analysis for Exp-concave Empirical Minimization with Arbitrary Convex Regularizer

In this paper, we present a simple analysis of {\bf fast rates} with {\it high probability} of {\bf empirical minimization} for {\it stochastic composite optimization} over a finite-dimensional bounded convex set with exponential concave loss functions and an arbitrary convex regularization. To the best of our knowledge, this result is the first of its kind. As a byproduct, we can directly obtain the fast rate with {\it high probability} for exponential concave empirical risk minimization with and without any convex regularization, which not only extends existing results of empirical risk minimization but also provides a unified framework for analyzing exponential concave empirical risk minimization with and without {\it any} convex regularization. Our proof is very simple only exploiting the covering number of a finite-dimensional bounded set and a concentration inequality of random vectors.

stat.ML↗

arXiv2017

Finding polynomial loop invariants for probabilistic programs

Quantitative loop invariants are an essential element in the verification of probabilistic programs. Recently, multivariate Lagrange interpolation has been applied to synthesizing polynomial invariants. In this paper, we propose an alternative approach. First, we fix a polynomial template as a candidate of a loop invariant. Using Stengle's Positivstellensatz and a transformation to a sum-of-squares problem, we find sufficient conditions on the coefficients. Then, we solve a semidefinite programming feasibility problem to synthesize the loop invariants. If the semidefinite program is unfeasible, we backtrack after increasing the degree of the template. Our approach is semi-complete in the sense that it will always lead us to a feasible solution if one exists and numerical errors are small. Experimental results show the efficiency of our approach.

cs.LO↗

arXiv2017

Distribution-based bisimulation for labelled Markov processes

In this paper we propose a (sub)distribution-based bisimulation for labelled Markov processes and compare it with earlier definitions of state and event bisimulation, which both only compare states. In contrast to those state-based bisimulations, our distribution bisimulation is weaker, but corresponds more closely to linear properties. We construct a logic and a metric to describe our distribution bisimulation and discuss linearity, continuity and compositional properties.

cs.LO↗

arXiv2017

A Richer Theory of Convex Constrained Optimization with Reduced Projections and Improved Rates

This paper focuses on convex constrained optimization problems, where the solution is subject to a convex inequality constraint. In particular, we aim at challenging problems for which both projection into the constrained domain and a linear optimization under the inequality constraint are time-consuming, which render both projected gradient methods and conditional gradient methods (a.k.a. the Frank-Wolfe algorithm) expensive. In this paper, we develop projection reduced optimization algorithms for both smooth and non-smooth optimization with improved convergence rates under a certain regularity condition of the constraint function. We first present a general theory of optimization with only one projection. Its application to smooth optimization with only one projection yields $O(1/ε)$ iteration complexity, which improves over the $O(1/ε^2)$ iteration complexity established before for non-smooth optimization and can be further reduced under strong convexity. Then we introduce a local error bound condition and develop faster algorithms for non-strongly convex optimization at the price of a logarithmic number of projections. In particular, we achieve an iteration complexity of $\widetilde O(1/ε^{2(1-θ)})$ for non-smooth optimization and $\widetilde O(1/ε^{1-θ})$ for smooth optimization, where $θ\in(0,1]$ appearing the local error bound condition characterizes the functional local growth rate around the optimal solutions. Novel applications in solving the constrained $\ell_1$ minimization problem and a positive semi-definite constrained distance metric learning problem demonstrate that the proposed algorithms achieve significant speed-up compared with previous algorithms.

math.OC↗

arXiv2017

Design of Lead-Free Inorganic Halide Perovskites for Solar Cells via Cation-Transmutation

Hybrid organic-inorganic halide perovskites with the prototype material of CH$_{3}$NH$_{3}$PbI$_{3}$ have recently attracted intense interest as low-cost and high-performance photovoltaic absorbers. Despite the high power conversion efficiency exceeding 20% achieved by their solar cells, two key issues -- the poor device stabilities associated with their intrinsic material instability and the toxicity due to water soluble Pb$^{2+}$ -- need to be resolved before large-scale commercialization. Here, we address these issues by exploiting the strategy of cation-transmutation to design stable inorganic Pb-free halide perovskites for solar cells. The idea is to convert two divalent Pb$^{2+}$ ions into one monovalent M$^{+}$ and one trivalent M$^{3+}$ ions, forming a rich class of quaternary halides in double-perovskite structure. We find through first-principles calculations this class of materials have good phase stability against decomposition and wide-range tunable optoelectronic properties. With photovoltaic-functionality-directed materials screening, we identify eleven optimal materials with intrinsic thermodynamic stability, suitable band gaps, small carrier effective masses, and low excitons binding energies as promising candidates to replace Pb-based photovoltaic absorbers in perovskite solar cells. The chemical trends of phase stabilities and electronic properties are also established for this class of materials, offering useful guidance for the development of perovskite solar cells fabricated with them.

cond-mat.mtrl-sci↗

arXiv2017

Cu-In Halide Perovskite solar absorbers

The long-term chemical instability and the presence of toxic Pb in otherwise stellar solar absorber APbX$_{3}$ have hindered their large-scale commercialization. Previously explored ways to achieve Pb-free halide perovskites involved replacing Pb$^{2+}$ with other similar M$^{2+}$ cations in ns$^2$ electron configuration, e.g., Sn$^{2+}$ or by Bi$^{3+}$ (plus Ag$^+$), but unfortunately this showed either poor stability (M = Sn) or weakly absorbing oversized indirect gaps (M = Bi), prompting concerns that perhaps stability and good optoelectronic properties might be contraindicated. Herein, we exploit the electronic structure underpinning of classic Cu[In,Ga]Se$_{2}$ (CIGS) chalcopyrite solar absorbers to design Pb-free halide perovskites by transmuting 2Pb to the pair [B$^{IB}$ + C$^{III}$]. The resulting group of double perovskites with formula A$_2$BCX$_6$ (A = K, Rb, Cs; B = Cu, Ag; C = Ga, In; X = Cl, Br, I) benefits from the ionic, yet narrow-gap character of halide perovskites, and at the same time borrows the advantage of the strong and rapidly rising Cu(d)/Se(p) $\rightarrow$ Ga/In(s/p) valence-to-conduction-band absorption spectra known from CIGS. This constitutes a new group of CuIn-based Halide Perovskite (CIHP). Our first-principles calculations guided by such design principles indicate that the CIHPs class has members with clear thermodynamic stability, showing rather strong direct-gap optical transitions, and manifesting a wide-range of tunable gap values (from zero to about 2.5 eV) and combination of light electron and heavy-light hole effective masses. Materials screening of candidate CHIPs then identifies the best-of-class Rb$_2$[CuIn]Cl$_6$, Rb$_2$[AgIn]Br$_6$ and Cs$_2$[AgIn]Br$_6$, having direct band gaps of 1.36, 1.46 and 1.50 eV, and a theoretical spectroscopic limited maximal efficiency comparable to chalcopyrites and CH$_3$NH$_3$PbI$_3$.

cond-mat.mtrl-sci↗

arXiv2017

Empirical Risk Minimization for Stochastic Convex Optimization: $O(1/n)$- and $O(1/n^2)$-type of Risk Bounds

Although there exist plentiful theories of empirical risk minimization (ERM) for supervised learning, current theoretical understandings of ERM for a related problem---stochastic convex optimization (SCO), are limited. In this work, we strengthen the realm of ERM for SCO by exploiting smoothness and strong convexity conditions to improve the risk bounds. First, we establish an $\widetilde{O}(d/n + \sqrt{F_*/n})$ risk bound when the random function is nonnegative, convex and smooth, and the expected function is Lipschitz continuous, where $d$ is the dimensionality of the problem, $n$ is the number of samples, and $F_*$ is the minimal risk. Thus, when $F_*$ is small we obtain an $\widetilde{O}(d/n)$ risk bound, which is analogous to the $\widetilde{O}(1/n)$ optimistic rate of ERM for supervised learning. Second, if the objective function is also $λ$-strongly convex, we prove an $\widetilde{O}(d/n + κF_*/n )$ risk bound where $κ$ is the condition number, and improve it to $O(1/[λn^2] + κF_*/n)$ when $n=\widetildeΩ(κd)$. As a result, we obtain an $O(κ/n^2)$ risk bound under the condition that $n$ is large and $F_*$ is small, which to the best of our knowledge, is the first $O(1/n^2)$-type of risk bound of ERM. Third, we stress that the above results are established in a unified framework, which allows us to derive new risk bounds under weaker conditions, e.g., without convexity of the random function and Lipschitz continuity of the expected function. Finally, we demonstrate that to achieve an $O(1/[λn^2] + κF_*/n)$ risk bound for supervised learning, the $\widetildeΩ(κd)$ requirement on $n$ can be replaced with $Ω(κ^2)$, which is dimensionality-independent.

cs.LG↗

arXiv2017

Construction of crystal structure prototype database: methods and applications

Crystal structure prototype data have become a useful source of information for materials discovery in the fields of crystallography, chemistry, physics, and materials science. This work reports the development of a robust and efficient method for assessing the similarity of structures on the basis of their interatomic distances. Using this method, we proposed a simple and unambiguous definition of crystal structure prototype based on hierarchical clustering theory, and constructed the Crystal Structure Prototype Database (CSPD) by filtering the known crystallographic structures in a database. With similar method, a program Structure Prototype Analysis Package (SPAP) was developed to remove similar structures in CALYPSO prediction results and extract predicted low energy structures for a separate theoretical structure database. A series of statistics describing the distribution of crystal structure prototypes in the CSPD was compiled to provide an important insight for structure prediction and high-throughput calculations. Illustrative examples of the application of the proposed database are given, including the generation of initial structures for structure prediction and determination of the prototype structure in databases. These examples demonstrate the CSPD to be a generally applicable and useful tool for materials discovery.

cond-mat.mtrl-sci↗

arXiv2017

A Novel Learning Algorithm for Büchi Automata based on Family of DFAs and Classification Trees

In this paper, we propose a novel algorithm to learn a Büchi automaton from a teacher who knows an $ω$-regular language. The algorithm is based on learning a formalism named family of DFAs (FDFAs) recently proposed by Angluin and Fisman[10]. The main catch is that we use a classification tree structure instead of the standard observation table structure. The worst case storage space required by our algorithm is quadratically better than the table-based algorithm proposed in [10]. We implement the first publicly available library ROLL (Regular Omega Language Learning ), which consists of all $ω$-regular learning algorithms available in the literature and the new algorithms proposed in this paper. Experimental results show that our tree-based algorithms have the best performance among others regarding the number of solved learning tasks.

cs.FL↗

arXiv2016

Efficient Non-oblivious Randomized Reduction for Risk Minimization with Improved Excess Risk Guarantee

In this paper, we address learning problems for high dimensional data. Previously, oblivious random projection based approaches that project high dimensional features onto a random subspace have been used in practice for tackling high-dimensionality challenge in machine learning. Recently, various non-oblivious randomized reduction methods have been developed and deployed for solving many numerical problems such as matrix product approximation, low-rank matrix approximation, etc. However, they are less explored for the machine learning tasks, e.g., classification. More seriously, the theoretical analysis of excess risk bounds for risk minimization, an important measure of generalization performance, has not been established for non-oblivious randomized reduction methods. It therefore remains an open problem what is the benefit of using them over previous oblivious random projection based approaches. To tackle these challenges, we propose an algorithmic framework for employing non-oblivious randomized reduction method for general empirical risk minimizing in machine learning tasks, where the original high-dimensional features are projected onto a random subspace that is derived from the data with a small matrix approximation error. We then derive the first excess risk bound for the proposed non-oblivious randomized reduction approach without requiring strong assumptions on the training data. The established excess risk bound exhibits that the proposed approach provides much better generalization performance and it also sheds more insights about different randomized reduction approaches. Finally, we conduct extensive experiments on both synthetic and real-world benchmark datasets, whose dimension scales to $O(10^7)$, to demonstrate the efficacy of our proposed approach.

cs.LG↗

arXiv2016

Functionality-directed Screening of Pb-free Hybrid Organic-inorganic Perovskites with Desired Intrinsic Photovoltaic Functionalities

The material class of hybrid organic-inorganic perovskites has risen rapidly from a virtually unknown material in photovoltaic applications a short 7 years ago into a ~20% efficient thin-film solar cell material. As promising as this class of materials is, however, there are limitations associated with its poor long-term stability, non-optimal band gap, presence of environmentally-toxic Pb element, etc. We herein apply a functionality-directed theoretical materials selection approach as a filter for initial screening of the compounds that satisfy the desired intrinsic photovoltaic functionalities and might overcome the above limitations. First-principles calculations are employed to systemically study thermodynamic stability and photovoltaic-related properties of hundred of candidate hybrid perovskites. We have identified in this materials selection process fourteen Ge and Sn-based materials with potential superior bulk-material-intrinsic photovoltaic performance. A distinct class of compounds containing NH3COH$^+$ with the organic molecule derived states intriguingly emerging at band-edges is found. Comparison of various candidate materials offers insights on how composition variation and microscopic structural changes affect key photovoltaic relevant properties in this family of materials.

cond-mat.mtrl-sci↗

arXiv2016

Sn(II)-containing phosphates as optoelectronic materials

We theoretically investigate Sn(II) phosphates as optoelectronic materials using first principles calculations. We focus on known prototype materials Sn$_n$P$_2$O$_{5+n}$ (n=2, 3, 4, 5) and a previously unreported compound, SnP$_2$O$_6$ (n=1), which we find using global optimization structure prediction. The electronic structure calculations indicate that these compounds all have large band gaps above 3.2 eV, meaning their transparency to visible light. Several of these compounds show relatively low hole effective masses ($\sim$2-3 m$_0$), comparable the electron masses. This suggests potential bipolar conductivity depending on doping. The dispersive valence band-edges underlying the low hole masses, originate from the anti-bonding hybridization between the Sn 5s orbitals and the phosphate groups. Analysis of structure-property relationships for the metastable structures generated during structure search shows considerable variation in combinations of band gap and carrier effective masses, implying chemical tunability of these properties. The unusual combinations of relatively high band gap, low carrier masses and high chemical stability suggests possible optoelectronic applications of these Sn(II) phosphates, including p-type transparent conductors. Related to this, calculations for doped material indicate low visible light absorption, combined with high plasma frequencies.

cond-mat.mtrl-sci↗

arXiv2016

Sparse Learning for Large-scale and High-dimensional Data: A Randomized Convex-concave Optimization Approach

In this paper, we develop a randomized algorithm and theory for learning a sparse model from large-scale and high-dimensional data, which is usually formulated as an empirical risk minimization problem with a sparsity-inducing regularizer. Under the assumption that there exists a (approximately) sparse solution with high classification accuracy, we argue that the dual solution is also sparse or approximately sparse. The fact that both primal and dual solutions are sparse motivates us to develop a randomized approach for a general convex-concave optimization problem. Specifically, the proposed approach combines the strength of random projection with that of sparse learning: it utilizes random projection to reduce the dimensionality, and introduces $\ell_1$-norm regularization to alleviate the approximation error caused by random projection. Theoretical analysis shows that under favored conditions, the randomized algorithm can accurately recover the optimal solutions to the convex-concave optimization problem (i.e., recover both the primal and dual solutions).

cs.LG↗

arXiv2016

A weighted pair graph representation for reconstructibility of Boolean control networks

A new concept of weighted pair graphs (WPGs) is proposed to represent a new reconstructibility definition for Boolean control networks (BCNs), which is a generalization of the reconstructibility definition given in [Fornasini & Valcher, TAC2013, Def. 4]. Based on the WPG representation, an effective algorithm for determining the new reconstructibility notion for BCNs is designed with the help of the theories of finite automata and formal languages. We prove that a BCN is not reconstructible iff its WPG has a complete subgraph. Besides, we prove that a BCN is reconstructible in the sense of [Fornasini & Valcher, TAC2013, Def. 4] iff its WPG has no cycles, which is simpler to be checked than the condition in [Fornasini & Valcher, TAC2013, Thm. 4].

math.OC↗

arXiv2016

A Simple Homotopy Proximal Mapping for Compressive Sensing

In this paper, we present a novel yet simple homotopy proximal mapping algorithm for compressive sensing. The algorithm adopts a simple proximal mapping of the $\ell_1$ norm at each iteration and gradually reduces the regularization parameter for the $\ell_1$ norm. We prove a global linear convergence of the proposed homotopy proximal mapping (HPM) algorithm for solving compressive sensing under three different settings (i) sparse signal recovery under noiseless measurements, (ii) sparse signal recovery under noisy measurements, and (iii) nearly-sparse signal recovery under sub-gaussian noisy measurements. In particular, we show that when the measurement matrix satisfies Restricted Isometric Properties (RIP), our theoretical results in settings (i) and (ii) almost recover the best condition on the RIP constants for compressive sensing. In addition, in setting (iii), our results for sparse signal recovery are better than the previous results, and furthermore our analysis explicitly exhibits that more observations lead to not only more accurate recovery but also faster convergence. Compared with previous studies on linear convergence for sparse signal recovery, our algorithm is simple and efficient, and our results are better and provide more insights. Finally our empirical studies provide further support for the proposed homotopy proximal mapping algorithm and verify the theoretical results.

cs.IT↗

arXiv2016

Verify LTL with Fairness Assumptions Efficiently

This paper deals with model checking problems with respect to LTL properties under fairness assumptions. We first present an efficient algorithm to deal with a fragment of fairness assumptions and then extend the algorithm to handle arbitrary %fairness assumptions ones. Notably, by making use of some syntactic transformations, our algorithm avoids to construct corresponding Büchi automata for the whole fairness assumptions, which can be very large in practice. We implement our algorithm in NuSMV and consider a large selection of formulas. Our experiments show that in many cases our approach exceeds the automata-theoretic approach up to several orders of magnitude, in both time and memory.

cs.LO↗

arXiv2016

Synthesising Strategy Improvement and Recursive Algorithms for Solving 2.5 Player Parity Games

2.5 player parity games combine the challenges posed by 2.5 player reachability games and the qualitative analysis of parity games. These two types of problems are best approached with different types of algorithms: strategy improvement algorithms for 2.5 player reachability games and recursive algorithms for the qualitative analysis of parity games. We present a method that - in contrast to existing techniques - tackles both aspects with the best suited approach and works exclusively on the 2.5 player game itself. The resulting technique is powerful enough to handle games with several million states.

cs.LO↗