TY - GEN
AB - We consider Markov decision processes (MDPs) with multiple limit-average (or mean-payoff) objectives.
There have been two different views: (i) the expectation semantics, where the goal is to optimize the expected mean-payoff objective, and (ii) the satisfaction semantics, where the goal is to maximize the probability of runs such that the mean-payoff value stays above a given vector.
We consider the problem where the goal is to optimize the expectation under the constraint that the satisfaction semantics is ensured, and thus consider a generalization that unifies the existing semantics. Our problem captures the notion of optimization with respect to strategies that are risk-averse (i.e., ensures certain probabilistic guarantee).
Our main results are algorithms for the decision problem which are always polynomial in the size of the MDP.
We also show that an approximation of the Pareto-curve can be computed in time polynomial in the size of the MDP, and the approximation factor, but exponential in the number of dimensions. Finally, we present a complete characterization of the strategy complexity (in terms of memory bounds and randomization) required to solve our problem.
AU - Chatterjee, Krishnendu
AU - Komarkova, Zuzana
AU - Kretinsky, Jan
ID - 5435
SN - 2664-1690
TI - Unifying two views on multiple mean-payoff objectives in Markov decision processes
ER -
TY - GEN
AB - Recently there has been a significant effort to handle quantitative properties in formal verification and synthesis. While weighted automata over finite and infinite words provide a natural and flexible framework to express quantitative properties, perhaps surprisingly, some basic system properties such as average response time cannot be expressed using weighted automata, nor in any other know decidable formalism. In this work, we introduce nested weighted automata as a natural extension of weighted automata which makes it possible to express important quantitative properties such as average response time.
In nested weighted automata, a master automaton spins off and collects results from weighted slave automata, each of which computes a quantity along a finite portion of an infinite word. Nested weighted automata can be viewed as the quantitative analogue of monitor automata, which are used in run-time verification. We establish an almost complete decidability picture for the basic decision problems about nested weighted automata, and illustrate their applicability in several domains. In particular, nested weighted automata can be used to decide average response time properties.
AU - Chatterjee, Krishnendu
AU - Henzinger, Thomas A
AU - Otop, Jan
ID - 5436
SN - 2664-1690
TI - Nested weighted automata
ER -
TY - GEN
AB - We consider the core algorithmic problems related to verification of systems with respect to three classical quantitative properties, namely, the mean-payoff property, the ratio property, and the minimum initial credit for energy property.
The algorithmic problem given a graph and a quantitative property asks to compute the optimal value (the infimum value over all traces) from every node of the graph. We consider graphs with constant treewidth, and it is well-known that the control-flow graphs of most programs have constant treewidth. Let $n$ denote the number of nodes of a graph, $m$ the number of edges (for constant treewidth graphs $m=O(n)$) and $W$ the largest absolute value of the weights.
Our main theoretical results are as follows.
First, for constant treewidth graphs we present an algorithm that approximates the mean-payoff value within a multiplicative factor of $\epsilon$ in time $O(n \cdot \log (n/\epsilon))$ and linear space, as compared to the classical algorithms that require quadratic time. Second, for the ratio property we present an algorithm that for constant treewidth graphs works in time $O(n \cdot \log (|a\cdot b|))=O(n\cdot\log (n\cdot W))$, when the output is $\frac{a}{b}$, as compared to the previously best known algorithm with running time $O(n^2 \cdot \log (n\cdot W))$. Third, for the minimum initial credit problem we show that (i)~for general graphs the problem can be solved in $O(n^2\cdot m)$ time and the associated decision problem can be solved in $O(n\cdot m)$ time, improving the previous known $O(n^3\cdot m\cdot \log (n\cdot W))$ and $O(n^2 \cdot m)$ bounds, respectively; and (ii)~for constant treewidth graphs we present an algorithm that requires $O(n\cdot \log n)$ time, improving the previous known $O(n^4 \cdot \log (n \cdot W))$ bound.
We have implemented some of our algorithms and show that they present a significant speedup on standard benchmarks.
AU - Chatterjee, Krishnendu
AU - Ibsen-Jensen, Rasmus
AU - Pavlogiannis, Andreas
ID - 5437
SN - 2664-1690
TI - Faster algorithms for quantitative verification in constant treewidth graphs
ER -
TY - GEN
AB - The edit distance between two words w1, w2 is the minimal number of word operations (letter insertions, deletions, and substitutions) necessary to transform w1 to w2. The edit distance generalizes to languages L1, L2, where the edit distance is the minimal number k such that for every word from L1 there exists a word in L2 with edit distance at most k. We study the edit distance computation problem between pushdown automata and their subclasses.
The problem of computing edit distance to a pushdown automaton is undecidable, and in practice, the interesting question is to compute the edit distance from a pushdown automaton (the implementation, a standard model for programs with recursion) to a regular language (the specification). In this work, we present a complete picture of decidability and complexity for deciding whether, for a given threshold k, the edit distance from a pushdown automaton to a finite automaton is at most k.
AU - Chatterjee, Krishnendu
AU - Henzinger, Thomas A
AU - Ibsen-Jensen, Rasmus
AU - Otop, Jan
ID - 5438
SN - 2664-1690
TI - Edit distance for pushdown automata
ER -
TY - GEN
AB - The target discounted-sum problem is the following: Given a rational discount factor 0 < λ < 1 and three rational values a, b, and t, does there exist a finite or an infinite sequence w ε(a, b)∗ or w ε(a, b)w, such that Σ|w| i=0 w(i)λi equals t? The problem turns out to relate to many fields of mathematics and computer science, and its decidability question is surprisingly hard to solve. We solve the finite version of the problem, and show the hardness of the infinite version, linking it to various areas and open problems in mathematics and computer science: β-expansions, discounted-sum automata, piecewise affine maps, and generalizations of the Cantor set. We provide some partial results to the infinite version, among which are solutions to its restriction to eventually-periodic sequences and to the cases that λ λ 1/2 or λ = 1/n, for every n ε N. We use our results for solving some open problems on discounted-sum automata, among which are the exact-value problem for nondeterministic automata over finite words and the universality and inclusion problems for functional automata.
AU - Boker, Udi
AU - Henzinger, Thomas A
AU - Otop, Jan
ID - 5439
SN - 2664-1690
TI - The target discounted-sum problem
ER -
TY - GEN
AB - Evolution occurs in populations of reproducing individuals. The structure of the population affects the outcome of the evolutionary process. Evolutionary graph theory is a powerful approach to study this phenomenon. There are two graphs. The interaction graph specifies who interacts with whom for payoff in the context of evolution. The replacement graph specifies who competes with whom for reproduction. The vertices of the two graphs are the same, and each vertex corresponds to an individual of the population. The fitness (or the reproductive rate) is a non-negative number, and depends on the payoff. A key quantity is the fixation probability of a new mutant. It is defined as the probability that a newly introduced mutant (on a single vertex) generates a lineage of offspring which eventually takes over the entire population of resident individuals. The basic computational questions are as follows: (i) the qualitative question asks whether the fixation probability is positive; and (ii) the quantitative approximation question asks for an approximation of the fixation probability. Our main results are as follows: First, we consider a special case of the general problem, where the residents do not reproduce. We show that the qualitative question is NP-complete, and the quantitative approximation question is #P-complete, and the hardness results hold even in the special case where the interaction and the replacement graphs coincide. Second, we show that in general both the qualitative and the quantitative approximation questions are PSPACE-complete. The PSPACE-hardness result for quantitative approximation holds even when the fitness is always positive.
AU - Chatterjee, Krishnendu
AU - Ibsen-Jensen, Rasmus
AU - Nowak, Martin
ID - 5440
SN - 2664-1690
TI - The complexity of evolutionary games on graphs
ER -
TY - GEN
AB - We study algorithmic questions for concurrent systems where the transitions are labeled from a complete, closed semiring, and path properties are algebraic with semiring operations. The algebraic path properties can model dataflow analysis problems, the shortest path problem, and many other natural problems that arise in program analysis. We consider that each component of the concurrent system is a graph with constant treewidth, a property satisfied by the controlflow graphs of most programs. We allow for multiple possible queries, which arise naturally in demand driven dataflow analysis. The study of multiple queries allows us to consider the tradeoff between the resource usage of the one-time preprocessing and for each individual query. The traditional approach constructs the product graph of all components and applies the best-known graph algorithm on the product. In this approach, even the answer to a single query requires the transitive closure (i.e., the results of all possible queries), which provides no room for tradeoff between preprocessing and query time. Our main contributions are algorithms that significantly improve the worst-case running time of the traditional approach, and provide various tradeoffs depending on the number of queries. For example, in a concurrent system of two components, the traditional approach requires hexic time in the worst case for answering one query as well as computing the transitive closure, whereas we show that with one-time preprocessing in almost cubic time, each subsequent query can be answered in at most linear time, and even the transitive closure can be computed in almost quartic time. Furthermore, we establish conditional optimality results showing that the worst-case running time of our algorithms cannot be improved without achieving major breakthroughs in graph algorithms (i.e., improving the worst-case bound for the shortest path problem in general graphs). Preliminary experimental results show that our algorithms perform favorably on several benchmarks.
AU - Chatterjee, Krishnendu
AU - Ibsen-Jensen, Rasmus
AU - Goharshady, Amir
AU - Pavlogiannis, Andreas
ID - 5441
SN - 2664-1690
TI - Algorithms for algebraic path properties in concurrent systems of constant treewidth components
ER -
TY - GEN
AB - We study algorithmic questions for concurrent systems where the transitions are labeled from a complete, closed semiring, and path properties are algebraic with semiring operations. The algebraic path properties can model dataflow analysis problems, the shortest path problem, and many other natural properties that arise in program analysis.
We consider that each component of the concurrent system is a graph with constant treewidth, and it is known that the controlflow graphs of most programs have constant treewidth. We allow for multiple possible queries, which arise naturally in demand driven dataflow analysis problems (e.g., alias analysis). The study of multiple queries allows us to consider the tradeoff between the resource usage of the \emph{one-time} preprocessing and for \emph{each individual} query. The traditional approaches construct the product graph of all components and apply the best-known graph algorithm on the product. In the traditional approach, even the answer to a single query requires the transitive closure computation (i.e., the results of all possible queries), which provides no room for tradeoff between preprocessing and query time.
Our main contributions are algorithms that significantly improve the worst-case running time of the traditional approach, and provide various tradeoffs depending on the number of queries. For example, in a concurrent system of two components, the traditional approach requires hexic time in the worst case for answering one query as well as computing the transitive closure, whereas we show that with one-time preprocessing in almost cubic time,
each subsequent query can be answered in at most linear time, and even the transitive closure can be computed in almost quartic time. Furthermore, we establish conditional optimality results that show that the worst-case running times of our algorithms cannot be improved without achieving major breakthroughs in graph algorithms (such as improving
the worst-case bounds for the shortest path problem in general graphs whose current best-known bound has not been improved in five decades). Finally, we provide a prototype implementation of our algorithms which significantly outperforms the existing algorithmic methods on several benchmarks.
AU - Anonymous, 1
AU - Anonymous, 2
AU - Anonymous, 3
AU - Anonymous, 4
ID - 5442
SN - 2664-1690
TI - Algorithms for algebraic path properties in concurrent systems of constant treewidth components
ER -
TY - GEN
AB - POMDPs are standard models for probabilistic planning problems, where an agent interacts with an uncertain environment. We study the problem of almost-sure reachability, where given a set of target states, the question is to decide whether there is a policy to ensure that the target set is reached with probability 1 (almost-surely). While in general the problem is EXPTIME-complete, in many practical cases policies with a small amount of memory suffice. Moreover, the existing solution to the problem is explicit, which first requires to construct explicitly an exponential reduction to a belief-support MDP. In this work, we first study the existence of observation-stationary strategies, which is NP-complete, and then small-memory strategies. We present a symbolic algorithm by an efficient encoding to SAT and using a SAT solver for the problem. We report experimental results demonstrating the scalability of our symbolic (SAT-based) approach.
AU - Chatterjee, Krishnendu
AU - Chmelik, Martin
AU - Davies, Jessica
ID - 5443
SN - 2664-1690
TI - A symbolic SAT-based algorithm for almost-sure reachability with small strategies in POMDPs
ER -
TY - GEN
AB - A comprehensive understanding of the clonal evolution of cancer is critical for understanding neoplasia. Genome-wide sequencing data enables evolutionary studies at unprecedented depth. However, classical phylogenetic methods often struggle with noisy sequencing data of impure DNA samples and fail to detect subclones that have different evolutionary trajectories. We have developed a tool, called Treeomics, that allows us to reconstruct the phylogeny of a cancer with commonly available sequencing technologies. Using Bayesian inference and Integer Linear Programming, robust phylogenies consistent with the biological processes underlying cancer evolution were obtained for pancreatic, ovarian, and prostate cancers. Furthermore, Treeomics correctly identified sequencing artifacts such as those resulting from low statistical power; nearly 7% of variants were misclassified by conventional statistical methods. These artifacts can skew phylogenies by creating illusory tumor heterogeneity among distinct samples. Importantly, we show that the evolutionary trees generated with Treeomics are mathematically optimal.
AU - Reiter, Johannes
AU - Makohon-Moore, Alvin
AU - Gerold, Jeffrey
AU - Bozic, Ivana
AU - Chatterjee, Krishnendu
AU - Iacobuzio-Donahue, Christine
AU - Vogelstein, Bert
AU - Nowak, Martin
ID - 5444
SN - 2664-1690
TI - Reconstructing robust phylogenies of metastatic cancers
ER -
TY - DATA
AB - This repository contains the experimental part of the CAV 2015 publication Counterexample Explanation by Learning Small Strategies in Markov Decision Processes.
We extended the probabilistic model checker PRISM to represent strategies of Markov Decision Processes as Decision Trees.
The archive contains a java executable version of the extended tool (prism_dectree.jar) together with a few examples of the PRISM benchmark library.
To execute the program, please have a look at the README.txt, which provides instructions and further information on the archive.
The archive contains scripts that (if run often enough) reproduces the data presented in the publication.
AU - Fellner, Andreas
ID - 5549
KW - Markov Decision Process
KW - Decision Tree
KW - Probabilistic Verification
KW - Counterexample Explanation
TI - Experimental part of CAV 2015 publication: Counterexample Explanation by Learning Small Strategies in Markov Decision Processes
ER -
TY - JOUR
AB - Parasitism creates selection for resistance mechanisms in host populations and is hypothesized to promote increased host evolvability. However, the influence of these traits on host evolution when parasites are no longer present is unclear. We used experimental evolution and whole-genome sequencing of Escherichia coli to determine the effects of past and present exposure to parasitic viruses (phages) on the spread of mutator alleles, resistance, and bacterial competitive fitness. We found that mutator alleles spread rapidly during adaptation to any of four different phage species, and this pattern was even more pronounced with multiple phages present simultaneously. However, hypermutability did not detectably accelerate adaptation in the absence of phages and recovery of fitness costs associated with resistance. Several lineages evolved phage resistance through elevated mucoidy, and during subsequent evolution in phage-free conditions they rapidly reverted to nonmucoid, phage-susceptible phenotypes. Genome sequencing revealed that this phenotypic reversion was achieved by additional genetic changes rather than by genotypic reversion of the initial resistance mutations. Insertion sequence (IS) elements played a key role in both the acquisition of resistance and adaptation in the absence of parasites; unlike single nucleotide polymorphisms, IS insertions were not more frequent in mutator lineages. Our results provide a genetic explanation for rapid reversion of mucoidy, a phenotype observed in other bacterial species including human pathogens. Moreover, this demonstrates that the types of genetic change underlying adaptation to fitness costs, and consequently the impact of evolvability mechanisms such as increased point-mutation rates, depend critically on the mechanism of resistance.
AU - Wielgoss, Sébastien
AU - Bergmiller, Tobias
AU - Bischofberger, Anna M.
AU - Hall, Alex R.
ID - 5749
IS - 3
JF - Molecular Biology and Evolution
SN - 0737-4038
TI - Adaptation to Parasites and Costs of Parasite Resistance in Mutator and Nonmutator Bacteria
VL - 33
ER -
TY - JOUR
AB - We present here the first integer-based algorithm for constructing a well-defined lattice sphere specified by integer radius and integer center. The algorithm evolves from a unique correspondence between the lattice points comprising the sphere and the distribution of sum of three square numbers in integer intervals. We characterize these intervals to derive a useful set of recurrences, which, in turn, aids in efficient computation. Each point of the lattice sphere is determined by resorting to only a few primitive operations in the integer domain. The symmetry of its quadraginta octants provides an added advantage by confining the computation to its prima quadraginta octant. Detailed theoretical analysis and experimental results have been furnished to demonstrate its simplicity and elegance.
AU - Biswas, Ranita
AU - Bhowmick, Partha
ID - 5804
IS - 4
JF - Theoretical Computer Science
SN - 0304-3975
TI - From prima quadraginta octant to lattice sphere through primitive integer operations
VL - 624
ER -
TY - JOUR
AU - Biswas, Ranita
AU - Bhowmick, Partha
ID - 5807
IS - 11
JF - Theoretical Computer Science
SN - 0304-3975
TI - On different topological classes of spherical geodesic paths and circles inZ3
VL - 605
ER -
TY - JOUR
AU - Biswas, Ranita
AU - Bhowmick, Partha
ID - 5808
IS - 6-8
JF - The Visual Computer
SN - 0178-2789
TI - Layer the sphere
VL - 31
ER -
TY - JOUR
AB - Transcription of eukaryotic protein-coding genes commences with the assembly of a conserved initiation complex, which consists of RNA polymerase II (Pol II) and the general transcription factors, at promoter DNA. After two decades of research, the structural basis of transcription initiation is emerging. Crystal structures of many components of the initiation complex have been resolved, and structural information on Pol II complexes with general transcription factors has recently been obtained. Although mechanistic details await elucidation, available data outline how Pol II cooperates with the general transcription factors to bind to and open promoter DNA, and how Pol II directs RNA synthesis and escapes from the promoter.
AU - Sainsbury, Sarah
AU - Bernecky, Carrie A
AU - Cramer, Patrick
ID - 594
IS - 3
JF - Nature Reviews Molecular Cell Biology
TI - Structural basis of transcription initiation by RNA polymerase II
VL - 16
ER -
TY - JOUR
AB - Carbon dioxide (CO2) gradients are ubiquitous and provide animals with information about their environment, such as the potential presence of prey or predators. The nematode Caenorhabditis elegans avoids elevated CO2, and previous work identified three neuron pairs called “BAG,” “AFD,” and “ASE” that respond to CO2 stimuli. Using in vivo Ca2+ imaging and behavioral analysis, we show that C. elegans can detect CO2 independently of these sensory pathways. Many of the C. elegans sensory neurons we examined, including the AWC olfactory neurons, the ASJ and ASK gustatory neurons, and the ASH and ADL nociceptors, respond to a rise in CO2 with a rise in Ca2+. In contrast, glial sheath cells harboring the sensory endings of C. elegans’ major chemosensory neurons exhibit strong and sustained decreases in Ca2+ in response to high CO2. Some of these CO2 responses appear to be cell intrinsic. Worms therefore may couple detection of CO2 to that of other cues at the earliest stages of sensory processing. We show that C. elegans persistently suppresses oviposition at high CO2. Hermaphrodite-specific neurons (HSNs), the executive neurons driving egg-laying, are tonically inhibited when CO2 is elevated. CO2 modulates the egg-laying system partly through the AWC olfactory neurons: High CO2 tonically activates AWC by a cGMP-dependent mechanism, and AWC output inhibits the HSNs. Our work shows that CO2 is a more complex sensory cue for C. elegans than previously thought, both in terms of behavior and neural circuitry.
AU - Fenk, Lorenz A.
AU - de Bono, Mario
ID - 6118
IS - 27
JF - Proceedings of the National Academy of Sciences
SN - 0027-8424
TI - Environmental CO2 inhibits Caenorhabditis elegans egg-laying by modulating olfactory neurons and evokes widespread changes in neural activity
VL - 112
ER -
TY - JOUR
AB - Brains organize behavior and physiology to optimize the response to threats or opportunities. We dissect how 21% O2, an indicator of surface exposure, reprograms C. elegans' global state, inducing sustained locomotory arousal and altering expression of neuropeptides, metabolic enzymes, and other non-neural genes. The URX O2-sensing neurons drive arousal at 21% O2 by tonically activating the RMG interneurons. Stimulating RMG is sufficient to switch behavioral state. Ablating the ASH, ADL, or ASK sensory neurons connected to RMG by gap junctions does not disrupt arousal. However, disrupting cation currents in these neurons curtails RMG neurosecretion and arousal. RMG signals high O2 by peptidergic secretion. Neuropeptide reporters reveal neural circuit state, as neurosecretion stimulates neuropeptide expression. Neural imaging in unrestrained animals shows that URX and RMG encode O2 concentration rather than behavior, while the activity of downstream interneurons such as AVB and AIY reflect both O2 levels and the behavior being executed.
AU - Laurent, Patrick
AU - Soltesz, Zoltan
AU - Nelson, Geoffrey M
AU - Chen, Changchun
AU - Arellano-Carbajal, Fausto
AU - Levy, Emmanuel
AU - de Bono, Mario
ID - 6120
JF - eLife
SN - 2050-084X
TI - Decoding a neural circuit controlling global animal state in C. elegans
VL - 4
ER -
TY - JOUR
AB - The osteoclast-associated receptor (OSCAR) is a collagen-binding immune receptor with important roles in dendritic cell maturation and activation of inflammatory monocytes as well as in osteoclastogenesis. The crystal structure of the OSCAR ectodomain is presented, both free and in complex with a consensus triple-helical peptide (THP). The structures revealed a collagen-binding site in each immunoglobulin-like domain (D1 and D2). The THP binds near a predicted collagen-binding groove in D1, but a more extensive interaction with D2 is facilitated by the unusually wide D1-D2 interdomain angle in OSCAR. Direct binding assays, combined with site-directed mutagenesis, confirm that the primary collagen-binding site in OSCAR resides in D2, in marked contrast to the related collagen receptors, glycoprotein VI (GPVI) and leukocyte-associated immunoglobulin-like receptor-1 (LAIR-1). Monomeric OSCAR D1D2 binds to the consensus THP with a KD of 28 µM measured in solution, but shows a higher affinity (KD 1.5 μM) when binding to a solid-phase THP, most likely due to an avidity effect. These data suggest a 2-stage model for the interaction of OSCAR with a collagen fibril, with transient, low-affinity interactions initiated by the membrane-distal D1, followed by firm adhesion to the primary binding site in D2.
AU - Zhou, Long
AU - Hinerman, J. M.
AU - Blaszczyk, M.
AU - Miller, J. L. C.
AU - Conrady, D. G.
AU - Barrow, A. D.
AU - Chirgadze, D. Y.
AU - Bihan, D.
AU - Farndale, R. W.
AU - Herr, A. B.
ID - 6507
IS - 5
JF - Blood
SN - 0006-4971
TI - Structural basis for collagen recognition by the immune receptor OSCAR
VL - 127
ER -
TY - JOUR
AB - Motivated by the significant performance gains which polar codes experience under successive cancellation list decoding, their scaling exponent is studied as a function of the list size. In particular, the error probability is fixed, and the tradeoff between the block length and back-off from capacity is analyzed. A lower bound is provided on the error probability under MAP decoding with list size L for any binary-input memoryless output-symmetric channel and for any class of linear codes such that their minimum distance is unbounded as the block length grows large. Then, it is shown that under MAP decoding, although the introduction of a list can significantly improve the involved constants, the scaling exponent itself, i.e., the speed at which capacity is approached, stays unaffected for any finite list size. In particular, this result applies to polar codes, since their minimum distance tends to infinity as the block length increases. A similar result is proved for genie-aided successive cancellation decoding when transmission takes place over the binary erasure channel, namely, the scaling exponent remains constant for any fixed number of helps from the genie. Note that since genie-aided successive cancellation decoding might be strictly worse than successive cancellation list decoding, the problem of establishing the scaling exponent of the latter remains open.
AU - Mondelli, Marco
AU - Hassani, Hamed
AU - Urbanke, Rudiger
ID - 6736
IS - 9
JF - IEEE Transactions on Information Theory
TI - Scaling exponent of list decoders with applications to polar codes
VL - 61
ER -