In multi-objective reinforcement learning (MORL) the agent is provided with multiple feedback signals when performing an action. These signals can be independent, complementary or conflicting. Hence, MORL is the process of learning policies that optimize multiple criteria simultaneously. In this abstract, we briefly describe our extensions to single-objective multi-armed bandits and reinforceme...

Considering a vector optimization problem to which properly efficient solutions are defined by using convex cone-monotone scalarization functions, we attach to it, by means of perturbation theory, new vector duals. When the primal problem, the scalarization function and the perturbation function are particularized, different dual vector problems are obtained, some of them already known in the l...

Scalarization approaches are the simplest methods for solving the multiobjective problems. The idea of scalarization is based on decomposition of multiobjective problems into single objective sub-problems. Every one of these sub-problems can be solved in a parallel manner since they are independent with each other. Hence, as a scalarization approach, systematically modification on the desirabil...

A general duality framework in convex multiobjective optimization is established using the scalarization with K-strongly increasing functions and the conjugate duality for composed convex cone-constrained optimization problems. Other scalarizations used in the literature arise as particular cases and the general duality is specialized for some of them, namely linear scalarization, maximum(-line...

In the current work, we have formulated the optimal bit-allocation problem for a scalable codec of images or videos as a constrained vector-valued optimization problem and demonstrated that there can be many optimal solutions, called Pareto optimal points. In practice, the Pareto points are derived via the weighted sum scalarization approach. An important question which arises is whether all th...

In this paper we extend naturally the scalarization proximal point method to solve multiobjective unconstrained minimization problems, proposed by Apolinario et al.[1], from Euclidean spaces to Hadamard manifolds for locally Lipschitz and quasiconvex vector objective functions. Moreover, we present a convergence analysis, under some mild assumptions on the multiobjective function, for two inexa...

Abstarct: The conventional equilibria problem found in many economics and network models is based on a scalar cost, or a single objective. Recently, equilibria problems based on a vector cost, or multicriteria, have received considerable attention. In this paper, we study a scalarization method for analyzing network equilibria problems with vector-valued cost function. The method is based on th...

the main purpose of this paper is to obtain sufficient conditions for existence of points of coincidence and common fixed points for three self mappings in $b$-metric spaces. next, we obtain cone $b$-metric version of these results by using a scalarization function. our results extend and generalize several well known comparable results in the existing literature.

The paper presents main features of the conic scalarization method in multiobjective optimization. The conic scalarization method guarantee to generate all proper efficient solutions and does not require any kind of cenvexity or boundedness conditions. In addition the preference and reference point information of the decision maker is taken into consideretion by this method. Also in this paper,...

Indicator-based evolutionary algorithms are amongst the best performing methods for solving multi-objective optimization (MOO) problems. In reinforcement learning (RL), introducing a quality indicator in an algorithm’s decision logic was not attempted before. In this paper, we propose a novel on-line multi-objective reinforcement learning (MORL) algorithm that uses the hypervolume indicator as ...