We consider the learning task of prediction of formation of core stable coalition structure in hedonic games based on agents noisy preferences. We have considered two cases: complete information (noisy preferences of all the agents are entirely known
) and partial information (noisy preferences over some coalitions are only known). We introduce a noise model that probabilistically scales the valuations of coalitions. The performance metric is the probability of our prediction conditioned on all or few noisy preferences of the agents be correct. The nature of our results is that this prediction probability is relatively low, including being zero, and rarely it is one. In the complete information two-agent model, in which each agent `retains or `inflates the values of its coalitions, we identify the expressions of the prediction probabilities in terms of the noise probability. We identify the interval of the noise probability such that the prediction probability is at least a user-given threshold. It turned out that, for some noisy games, the noise probability interval does not exist for a threshold as low as 0.1481, thus demonstrating that the prediction probabilities are generally low even in this model. In the partial information setup, we consider $n$ agent games with $l$ support of noise values, and such noisy preferences are available for some coalitions only. We obtain the bounds on the prediction probability of a partition to be $epsilon$-PAC stable in the noise-free game in the cases when the realized noisy game has or hasnt $epsilon$-PAC stable outcome.
Both single-agent and multi-agent actor-critic algorithms are an important class of Reinforcement Learning algorithms. In this work, we propose three fully decentralized multi-agent natural actor-critic (MAN) algorithms. The agents objective is to co
llectively learn a joint policy that maximizes the sum of averaged long-term returns of these agents. In the absence of a central controller, agents communicate the information to their neighbors via a time-varying communication network while preserving privacy. We prove the convergence of all the 3 MAN algorithms to a globally asymptotically stable point of the ODE corresponding to the actor update; these use linear function approximations. We use the Fisher information matrix to obtain the natural gradients. The Fisher information matrix captures the curvature of the Kullback-Leibler (KL) divergence between polices at successive iterates. We also show that the gradient of this KL divergence between policies of successive iterates is proportional to the objective functions gradient. Our MAN algorithms indeed use this emph{representation} of the objective functions gradient. Under certain conditions on the Fisher information matrix, we prove that at each iterate, the optimal value via MAN algorithms can be better than that of the multi-agent actor-critic (MAAC) algorithm using the standard gradients. To validate the usefulness of our proposed algorithms, we implement all the 3 MAN algorithms on a bi-lane traffic network to reduce the average network congestion. We observe an almost 25% reduction in the average congestion in 2 MAN algorithms; the average congestion in another MAN algorithm is on par with the MAAC algorithm. We also consider a generic 15 agent MARL; the performance of the MAN algorithms is again as good as the MAAC algorithm. We attribute the better performance of the MAN algorithms to their use of the above representation.
Thompson Sampling has generated significant interest due to its better empirical performance than upper confidence bound based algorithms. In this paper, we study Thompson Sampling based algorithm for Unsupervised Sequential Selection (USS) problem.
The USS problem is a variant of the stochastic multi-armed bandits problem, where the loss of an arm can not be inferred from the observed feedback. In the USS setup, arms are associated with fixed costs and are ordered, forming a cascade. In each round, the learner selects an arm and observes the feedback from arms up to the selected arm. The learners goal is to find the arm that minimizes the expected total loss. The total loss is the sum of the cost incurred for selecting the arm and the stochastic loss associated with the selected arm. The problem is challenging because, without knowing the mean loss, one cannot compute the total loss for the selected arm. Clearly, learning is feasible only if the optimal arm can be inferred from the problem structure. As shown in the prior work, learning is possible when the problem instance satisfies the so-called `Weak Dominance (WD) property. Under WD, we show that our Thompson Sampling based algorithm for the USS problem achieves near optimal regret and has better numerical performance than existing algorithms.
We investigate optimal posteriors for recently introduced cite{begin2016pac} chi-squared divergence based PAC-Bayesian bounds in terms of nature of their distribution, scalability of computations, and test set performance. For a finite classifier set
, we deduce bounds for three distance functions: KL-divergence, linear and squared distances. Optimal posterior weights are proportional to deviations of empirical risks, usually with subset support. For uniform prior, it is sufficient to search among posteriors on classifier subsets ordered by these risks. We show the bound minimization for linear distance as a convex program and obtain a closed-form expression for its optimal posterior. Whereas that for squared distance is a quasi-convex program under a specific condition, and the one for KL-divergence is non-convex optimization (a difference of convex functions). To compute such optimal posteriors, we derive fast converging fixed point (FP) equations. We apply these approaches to a finite set of SVM regularization parameter values to yield stochastic SVMs with tight bounds. We perform a comprehensive performance comparison between our optimal posteriors and known KL-divergence based posteriors on a variety of UCI datasets with varying ranges and variances in risk values, etc. Chi-squared divergence based posteriors have weaker bounds and worse test errors, hinting at an underlying regularization by KL-divergence based posteriors. Our study highlights the impact of divergence function on the performance of PAC-Bayesian classifiers. We compare our stochastic classifiers with cross-validation based deterministic classifier. The latter has better test errors, but ours is more sample robust, has quantifiable generalization guarantees, and is computationally much faster.
PAC-Bayesian set up involves a stochastic classifier characterized by a posterior distribution on a classifier set, offers a high probability bound on its averaged true risk and is robust to the training sample used. For a given posterior, this bound
captures the trade off between averaged empirical risk and KL-divergence based model complexity term. Our goal is to identify an optimal posterior with the least PAC-Bayesian bound. We consider a finite classifier set and 5 distance functions: KL-divergence, its Pinskers and a sixth degree polynomial approximations; linear and squared distances. Linear distance based model results in a convex optimization problem. We obtain closed form expression for its optimal posterior. For uniform prior, this posterior has full support with weights negative-exponentially proportional to number of misclassifications. Squared distance and Pinskers approximation bounds are possibly quasi-convex and are observed to have single local minimum. We derive fixed point equations (FPEs) using partial KKT system with strict positivity constraints. This obviates the combinatorial search for subset support of the optimal posterior. For uniform prior, exponential search on a full-dimensional simplex can be limited to an ordered subset of classifiers with increasing empirical risk values. These FPEs converge rapidly to a stationary point, even for a large classifier set when a solver fails. We apply these approaches to SVMs generated using a finite set of SVM regularization parameter values on 9 UCI datasets. These posteriors yield stochastic SVM classifiers with tight bounds. KL-divergence based bound is the tightest, but is computationally expensive due to non-convexity and multiple calls to a root finding algorithm. Optimal posteriors for all 5 distance functions have lowest 10% test error values on most datasets, with linear distance being the easiest to obtain.
