Robust learning in expert networks: a comparative analysis

Human experts as well as autonomous agents in a referral network must decide whether to accept a task or refer to a more appropriate expert, and if so to whom. In order for the referral network to improve over time, the experts must learn to estimate the topical expertise of other experts. This article extends concepts from Multi-agent Reinforcement Learning and Active Learning to referral networks for distributed learning in referral networks. Among a wide array of algorithms evaluated, Distributed Interval Estimation Learning (DIEL), based on Interval Estimation Learning, was found to be superior for learning appropriate referral choices, compared to ??-Greedy, Q-learning, Thompson Sampling and Upper Confidence Bound (UCB) methods. In addition to a synthetic data set, we compare the performance of the stronger learning-to-refer algorithms on a referral network of high-performance Stochastic Local Search (SLS) SAT solvers where expertise does not obey any known parameterized distribution. An evaluation of overall network performance and a robustness analysis is conducted across the learning algorithms, with an emphasis on capacity constraints and evolving networks, where experts with known expertise drop off and new experts of unknown performance enter — situations that arise in real-world scenarios but were heretofore ignored.     

Another Sub-exponential Algorithm for the Simple Stochastic Game

We study the problem of solving simple stochastic games, and give both an interesting new algorithm and a hardness result. We show a reduction from fine approximation of simple stochastic games to coarse approximation of a polynomial sized game, which can be viewed as an evidence showing the hardness to approximate the value of simple stochastic games. We also present a randomized algorithm that runs in \(\tilde{O}(\sqrt{|V_{\mathrm{R}}|!})\) time, where |V _R| is the number of RANDOM vertices and \(\tilde{O}\) ignores polynomial terms. This algorithm is the fastest known algorithm when |V _R|=ω(log?n) and \(|V_{\mathrm{R}}|=o(\sqrt{\min{|V_{\min}|,|V_{\max}|}})\) and it works for general (non-stopping) simple stochastic games.     

Time versus cost tradeoffs for deterministic rendezvous in networks

Two mobile agents, starting from different nodes of a network at possibly different times, have to meet at the same node. This problem is known as rendezvous. Agents move in synchronous rounds. Each agent has a distinct integer label from the set \(\{1,\ldots ,L\}\). Two main efficiency measures of rendezvous are its time (the number of rounds until the meeting) and its cost (the total number of edge traversals). We investigate tradeoffs between these two measures. A natural benchmark for both time and cost of rendezvous in a network is the number of edge traversals needed for visiting all nodes of the network, called the exploration time. Hence we express the time and cost of rendezvous as functions of an upper bound E on the time of exploration (where E and a corresponding exploration procedure are known to both agents) and of the size L of the label space. We present two natural rendezvous algorithms. Algorithm Cheap has cost O(E) (and, in fact, a version of this algorithm for the model where the agents start simultaneously has cost exactly E) and time O(EL). Algorithm Fast has both time and cost \(O(E\log L)\). Our main contributions are lower bounds showing that, perhaps surprisingly, these two algorithms capture the tradeoffs between time and cost of rendezvous almost tightly. We show that any deterministic rendezvous algorithm of cost asymptotically E (i.e., of cost \(E+o(E)\)) must have time \(\varOmega (EL)\). On the other hand, we show that any deterministic rendezvous algorithm with time complexity \(O(E\log L)\) must have cost \(\varOmega (E\log L)\).     

Power and welfare in bargaining for coalition structure formation

We investigate a noncooperative bargaining game for partitioning n agents into non-overlapping coalitions. The game has n time periods during which the players are called according to an exogenous agenda to propose offers. With probability \(\delta \), the game ends during any time period \(t<n\). If it does, the first t players on the agenda get a chance to propose but the others do not. Thus, \(\delta \) is a measure of the degree of democracy within the game (ranging from democracy for \(\delta =0\), through increasing levels of authoritarianism as \(\delta \) approaches 1, to dictatorship for \(\delta =1\)). We determine the subgame perfect equilibrium (SPE) and study how a player’s position on the agenda affects his bargaining power. We analyze the relation between the distribution of power of individual players, the level of democracy, and the welfare efficiency of the game. We find that purely democratic games are welfare inefficient and that introducing a degree of authoritarianism into the game makes the distribution of power more equitable and also maximizes welfare. These results remain invariant under two types of player preferences: one where each player’s preference is a total order on the space of possible coalition structures and the other where each player either likes or dislikes a coalition structure. Finally, we show that the SPE partition may or may not be core stable.     

Improving configuration checking for satisfiable random <Emphasis Type="Italic">k</Emphasis>-SAT instances

Local search algorithms based on the Configuration Checking (CC) strategy have been shown to be efficient in solving satisfiable random k-SAT instances. The purpose of the CC strategy is to avoid the cycling problem, which corresponds to revisiting already flipped variables too soon. It is done by considering the neighborhood of the formula variables. In this paper, we propose to improve the CC strategy on the basis of an empirical study of a powerful algorithm using this strategy. The first improvement introduces a new and simple criterion, which refines the selection of the variables to flip for the 3-SAT instances. The second improvement is achieved by using the powerful local search algorithm Novelty with the adaptive noise setting. This algorithm enhances the efficiency of the intensification and diversification phases when solving k-SAT instances with k ≥ 4. We name the resulting local search algorithm Ncca+ and show its effectiveness when treating satisfiable random k-SAT instances issued from the SAT Challenge 2012. Ncca+ won the bronze medal of the random SAT track of the SAT Competition 2013.     

Real-time scheduling algorithm for safety-critical systems on faulty multicore environments

An algorithm (called FTM) for scheduling of real-time sporadic tasks on a multicore platform is proposed. Each task has a deadline by which it must complete its non-erroneous execution. The FTM algorithm executes backups in order to recover from errors caused by non-permanent and permanent hardware faults. The worst-case schedulability analysis of FTM algorithm is presented considering an application-level error model, which is independent of the stochastic behavior of the underlying hardware-level fault model. Then, the stochastic behavior of hardware-level fault model is plugged in to the analysis to derive the probability of meeting all the deadlines. Such probabilistic guarantee is the level of assurance (i.e., reliability) regarding the correct functional and timing behaviors of the system. One of the salient features of FTM algorithm is that it executes some backups in active redundancy to exploit the parallel multicore architecture while other backups passively to avoid unnecessary execution of too many active backups. This paper also proposes a scheme to determine for each task the number of backups that should run in active redundancy in order to increase the probability of meeting all the deadlines. The effectiveness of the proposed approach is demonstrated using an example application.     

Validating gameplay activity inventory (GAIN) for modeling player profiles

In the present study, we validated Gameplay Activity Inventory (GAIN), a short and psychometrically sound instrument for measuring players’ gameplay preferences and modeling player profiles. In Study 1, participants in Finland (\(N=879\)) responded to a 52-item version of GAIN. An exploratory factor analysis was used to identify five latent factors of gameplay activity appreciation: Aggression, Management, Exploration, Coordination, and Caretaking. In Study 2, respondents in Canada (\(N=1322\)) and Japan (\(N=1178\)) responded to GAIN, and the factor structure of a 15-item version was examined using a Confirmatory Factor Analysis. The results showed that the short version of GAIN has good construct validity, convergent validity, and discriminant validity in Japan and in Canada. We demonstrated the usefulness of GAIN by conducting a cluster analysis to identify player types that differ in both demographics and game choice. GAIN can be used in research as a tool for investigating player profiles. Game companies, publishers and analysts can utilize GAIN in player-centric game development and targeted marketing and in generating personalized game recommendations.     

ReFlO: an interactive tool for pipe-and-filter domain specification and program generation

ReFlO is a framework and interactive tool to record and systematize domain knowledge used by experts to derive complex pipe-and-filter (PnF) applications. Domain knowledge is encoded as transformations that alter PnF graphs by refinement (adding more details), flattening (removing modular boundaries), and optimization (substituting inefficient PnF graphs with more efficient ones). All three kinds of transformations arise in reverse-engineering legacy PnF applications. We present the conceptual foundation and tool capabilities of ReFlO, illustrate how parallel PnF applications are designed and generated, and how domain-specific libraries of transformations are developed.     

Rendezvous in networks in spite of delay faults

Two mobile agents, starting from different nodes of an unknown network, have to meet at a node. Agents move in synchronous rounds using a deterministic algorithm. Each agent has a different label, which it can use in the execution of the algorithm, but it does not know the label of the other agent. Agents do not know any bound on the size of the network. In each round an agent decides if it remains idle or if it wants to move to one of the adjacent nodes. Agents are subject to delay faults: if an agent incurs a fault in a given round, it remains in the current node, regardless of its decision. If it planned to move and the fault happened, the agent is aware of it. We consider three scenarios of fault distribution: random (independently in each round and for each agent with constant probability \(0<p<1\)), unbounded adversarial (the adversary can delay an agent for an arbitrary finite number of consecutive rounds) and bounded adversarial (the adversary can delay an agent for at most c consecutive rounds, where c is unknown to the agents). The quality measure of a rendezvous algorithm is its cost, which is the total number of edge traversals. For random faults, we show an algorithm with cost polynomial in the size n of the network and polylogarithmic in the larger label L, which achieves rendezvous with very high probability in arbitrary networks. By contrast, for unbounded adversarial faults we show that rendezvous is not possible, even in the class of rings. Under this scenario we give a rendezvous algorithm with cost \(O(n\ell )\), where \(\ell \) is the smaller label, working in arbitrary trees, and we show that \(\varOmega (\ell )\) is the lower bound on rendezvous cost, even for the two-node tree. For bounded adversarial faults, we give a rendezvous algorithm working for arbitrary networks, with cost polynomial in n, and logarithmic in the bound c and in the larger label L.     

Endogenous games with goals: side-payments among goal-directed agents

Boolean games have been developed as a paradigm for modelling societies of goal-directed agents. In boolean games agents exercise control over propositional variables and strive to achieve a goal formula whose realization might require the opponents’ cooperation. The presence of agents that are goal-directed makes it difficult for an external authority to be able to remove undesirable properties that are inconsistent with agents’ goals, as shown by recent contributions in the multi-agent literature. What this paper does is to analyse the problem of regulation of goal-direct agents from within the system, i.e., what happens when agents themselves are given the chance to negotiate the strategies to be played with one another. Concretely, we introduce endogenous games with goals, obtained coupling a general model of goal-directed agents (strategic games with goals) with a general model of pre-play negotiations (endogenous games) coming from game theory. Strategic games with goals are shown to have a direct correspondence with strategic games (Proposition 1) but, when side-payments are allowed in the pre-play phase, display a striking imbalance (Proposition 4). The effect of side-payments can be fully simulated by taxation mechanisms studied in the literature (Proposition 7), yet we show sufficient conditions under which outcomes can be rationally sustained without external intervention (Proposition 5). Also, integrating taxation mechanisms and side-payments, we are able to transform our starting models in such a way that outcomes that are theoretically sustainable thanks to a pre-play phase can be actually sustained even with limited resources (Proposition 8). Finally, we show how an external authority incentivising a group of agents can be studied as a special agent of an appropriately extended endogenous game with goals (Proposition 11).     

Cheap talk,cooperation, and trust in global software engineering

In our studies of global software engineering (GSE) teams, we found that informal, non-work-related conversations are positively associated with trust. Seeking to use novel analytical techniques to more carefully investigate this phenomenon, we described these non-work-related conversations by adapting the economics literature concept of “cheap talk,” and studied it using Evolutionary Game Theory (EGT). More specifically, we modified the classic Stag-hunt game and analyzed the dynamics in a fixed population setting (an abstraction of a GSE team). Doing so, we were able to demonstrate how cheap talk over the Internet (e-cheap talk) was powerful enough to facilitate the emergence of trust and improve the probability of cooperation where the punishment for uncooperative behavior is comparable to the cost of the cheap talk. To validate the results of our theoretical approach, we conducted two empirical case studies that analyzed the logged IRC development discussions of Apache Lucene (http://lucene.apache. org/) and Chromium OS (http://www.chromium.org/chromium-os) using both quantitative and qualitative methods. The results provide general support to the theoretical propositions. We discuss our findings and the theoretical and practical implications to GSE collaborations and research.     

On Optimally Partitioning a Text to Improve Its Compression

In this paper we investigate the problem of partitioning an input string T in such a way that compressing individually its parts via a base-compressor C gets a compressed output that is shorter than applying C over the entire T at once. This problem was introduced in Buchsbaum et al. (Proc. of 11th ACM-SIAM Symposium on Discrete Algorithms, pp. 175–184, 2000; J. ACM 50(6):825–851, 2003) in the context of table compression, and then further elaborated and extended to strings and trees by Ferragina et al. (J. ACM 52:688–713, 2005; Proc. of 46th IEEE Symposium on Foundations of Computer Science, pp. 184–193, 2005) and Mäkinen and Navarro (Proc. of 14th Symposium on String Processing and Information Retrieval, pp. 229–241, 2007). Unfortunately, the literature offers poor solutions: namely, we know either a cubic-time algorithm for computing the optimal partition based on dynamic programming (Buchsbaum et al. in J. ACM 50(6):825–851, 2003; Giancarlo and Sciortino in Proc. of 14th Symposium on Combinatorial Pattern Matching, pp. 129–143, 2003), or few heuristics that do not guarantee any bounds on the efficacy of their computed partition (Buchsbaum et al. in Proc. of 11th ACM-SIAM Symposium on Discrete Algorithms, pp. 175–184, 2000; J. ACM 50(6):825–851, 2003), or algorithms that are efficient but work in some specific scenarios (such as the Burrows-Wheeler Transform, see e.g. Ferragina et al. in J. ACM 52:688–713, 2005; Mäkinen and Navarro in Proc. of 14th Symposium on String Processing and Information Retrieval, pp. 229–241, 2007) and achieve compression performance that might be worse than the optimal-partitioning by a Ω(log?n/log?log?n) factor. Therefore, computing efficiently the optimal solution is still open (Buchsbaum and Giancarlo in Encyclopedia of Algorithms, pp. 939–942, 2008). In this paper we provide the first algorithm which computes in O(nlog?_1+ε n) time and O(n) space, a partition of T whose compressed output is guaranteed to be no more than (1+ε)-worse the optimal one, where ε may be any positive constant fixed in advance. This result holds for any base-compressor C whose compression performance can be bounded in terms of the zero-th or the k-th order empirical entropy of the text T. We will also discuss extensions of our results to BWT-based compressors and to the compression booster of Ferragina et al. (J. ACM 52:688–713, 2005).     

Improved Approximation Algorithms for Label Cover Problems

In this paper we consider both the maximization variant Max Rep and the minimization variant Min Rep of the famous Label Cover problem. So far the best approximation ratios known for these two problems were \(O(\sqrt{n})\) and indeed some authors suggested the possibility that this ratio is the best approximation factor for these two problems. We show, in fact, that there are a O(n ^1/3)-approximation algorithm for Max Rep and a O(n ^1/3log?^2/3 n)-approximation algorithm for Min Rep. In addition, we also exhibit a randomized reduction from Densest k-Subgraph to Max Rep, showing that any approximation factor for Max Rep implies the same factor (up to a constant) for Densest k-Subgraph.     

Automated competitive analysis of real-time scheduling with graph games

This paper is devoted to automatic competitive analysis of real-time scheduling algorithms for firm-deadline tasksets, where only completed tasks contribute some utility to the system. Given such a taskset \({\mathcal {T}}\), the competitive ratio of an on-line scheduling algorithm \({\mathcal {A}}\) for \({\mathcal {T}}\) is the worst-case utility ratio of \({\mathcal {A}}\) over the utility achieved by a clairvoyant algorithm. We leverage the theory of quantitative graph games to address the competitive analysis and competitive synthesis problems. For the competitive analysis case, given any taskset \({\mathcal {T}}\) and any finite-memory on-line scheduling algorithm \({\mathcal {A}}\), we show that the competitive ratio of \({\mathcal {A}}\) in \({\mathcal {T}}\) can be computed in polynomial time in the size of the state space of \({\mathcal {A}}\). Our approach is flexible as it also provides ways to model meaningful constraints on the released task sequences that determine the competitive ratio. We provide an experimental study of many well-known on-line scheduling algorithms, which demonstrates the feasibility of our competitive analysis approach that effectively replaces human ingenuity (required for finding worst-case scenarios) by computing power. For the competitive synthesis case, we are just given a taskset \({\mathcal {T}}\), and the goal is to automatically synthesize an optimal on-line scheduling algorithm \({\mathcal {A}}\), i.e., one that guarantees the largest competitive ratio possible for \({\mathcal {T}}\). We show how the competitive synthesis problem can be reduced to a two-player graph game with partial information, and establish that the computational complexity of solving this game is Np-complete. The competitive synthesis problem is hence in Np in the size of the state space of the non-deterministic labeled transition system encoding the taskset. Overall, the proposed framework assists in the selection of suitable scheduling algorithms for a given taskset, which is in fact the most common situation in real-time systems design.     

A counterexample to the chain rule for conditional HILL entropy

Most entropy notions \({H(.)}\) like Shannon or min-entropy satisfy a chain rule stating that for random variables \({X,Z,}\) and \({A}\) we have \({H(X|Z,A)\ge H(X|Z)-|A|}\). That is, by conditioning on \({A}\) the entropy of \({X}\) can decrease by at most the bitlength \({|A|}\) of \({A}\). Such chain rules are known to hold for some computational entropy notions like Yao’s and unpredictability-entropy. For HILL entropy, the computational analogue of min-entropy, the chain rule is of special interest and has found many applications, including leakage-resilient cryptography, deterministic encryption, and memory delegation. These applications rely on restricted special cases of the chain rule. Whether the chain rule for conditional HILL entropy holds in general was an open problem for which we give a strong negative answer: we construct joint distributions \({(X,Z,A)}\), where \({A}\) is a distribution over a single bit, such that the HILL entropy H ^HILL \({(X|Z)}\) is large but H ^HILL \({(X|Z,A)}\) is basically zero.Our counterexample just makes the minimal assumption that \({{\mathbf{NP}} \nsubseteq{\mathbf{P/poly}}}\). Under the stronger assumption that injective one-way function exist, we can make all the distributions efficiently samplable.Finally, we show that some more sophisticated cryptographic objects like lossy functions can be used to sample a distribution constituting a counterexample to the chain rule making only a single invocation to the underlying object.     

Quantum secure two-party computation for set intersection with rational players

Recently, Shi et al. (Phys Rev A 92:022309, 2015) proposed quantum oblivious set member decision protocol where two legitimate parties, namely Alice and Bob, play a game. Alice has a secret k, and Bob has a set \(\{k_1,k_2,\ldots k_n\}\). The game is designed towards testing if the secret k is a member of the set possessed by Bob without revealing the identity of k. The output of the game will be either “Yes” (bit 1) or “No” (bit 0) and is generated at Bob’s place. Bob does not know the identity of k, and Alice does not know any element of the set. In a subsequent work (Shi et al in Quant Inf Process 15:363–371, 2016), the authors proposed a quantum scheme for private set intersection (PSI) where the client (Alice) gets the intersected elements with the help of a server (Bob) and the server knows nothing. In the present draft, we extended the game to compute the intersection of two computationally indistinguishable sets X and Y possessed by Alice and Bob, respectively. We consider Alice and Bob as rational players, i.e. they are neither “good” nor “bad”. They participate in the game towards maximizing their utilities. We prove that in this rational setting, the strategy profile ((cooperate, abort), (cooperate, abort)) is a strict Nash equilibrium. If ((cooperate, abort), (cooperate, abort)) is strict Nash, then fairness and correctness of the protocol are guaranteed.     

Quantum algorithm to solve function inversion with time–space trade-off

In general, it is a difficult problem to solve the inverse of any function. With the inverse implication operation, we present a quantum algorithm for solving the inversion of function via using time–space trade-off in this paper. The details are as follows. Let function \(f(x)=y\) have k solutions, where \(x\in \{0, 1\}^{n}, y\in \{0, 1\}^{m}\) for any integers n, m. We show that an iterative algorithm can be used to solve the inverse of function f(x) with successful probability \(1-\left( 1-\frac{k}{2^{n}}\right) ^{L}\) for \(L\in Z^{+}\). The space complexity of proposed quantum iterative algorithm is O(Ln), where L is the number of iterations. The paper concludes that, via using time–space trade-off strategy, we improve the successful probability of algorithm.     

An optimal adaptive stabilizing algorithm for two edge-disjoint paths problem

The problem of two edge-disjoint paths is to identify two paths \(Q_1\) and \(Q_2\) from source \(s \in V\) to target \(t \in V\) without any common arc in a directed connected graph \(G=(V, E)\). In this paper, we present an adaptive stabilizing algorithm for finding a pair of edge-disjoint paths from s to t in G in O(D) rounds with state-space complexity of \(O(log\; n)\) bits per process, where n is the number of nodes and D is the diameter of the graph. The proposed algorithm is optimal with respect to its time complexity, and the total length of the shortest paths. In addition, it can also be used to solve the problem for undirected graphs. Since the proposed algorithm is stabilizing, it does not require initialization and is capable of withstanding transient faults. We view a fault that perturbs the state of the system but not its program as a transient fault. In addition, the proposed algorithm is adaptive since it is capable of dealing with topology changes in the form of addition/removal of arcs and/or nodes as well as changes in the directions of arcs provided that two edge-disjoint paths between s and t exist after the topology change.     

Virtual reality software package for implementing motor learning and rehabilitation experiments

Virtual reality games for rehabilitation are attracting increasing growth. In particular, there is a demand for games that allow therapists to identify an individual’s difficulties and customize the control of variables, such as speed, size, distance, as well as visual and auditory feedback. This study presents and describes a virtual reality software package (Bridge Games) to promote rehabilitation of individuals living with disabilities and highlights preliminary researches of its use for implementing motor learning and rehabilitation. First, the study presents seven games in the software package that can be chosen by the rehabilitation team, considering the patient’s needs. All game characteristics are described including name, function presentation, objective and valuable measurements for rehabilitation. Second, preliminary results illustrate some applications of two games, considering 343 people with various disabilities and health status. Based on the results, in the Coincident Timing game, there was a main effect of movement sensor type (in this instance the most functional device was the keyboard when compared with Kinect and touch screen) on average time reached by sample analyzed, F(2, 225) = 4.42, p < 0.05. Similarly, in the Challenge! game, a main effect was found for movement sensor type. However, in this case, touch screen provided better performance than Kinect and Leap Motion, F(2, 709) = 5.90, p < 0.01. Thus, Bridge Games is a possible software game to quantify motor learning. Moreover, the findings suggest that motor skills might be practiced differently depending on the environmental interface in which the game may be used.