WorksComputer Science
Learning Global Equilibria in Non-Concave Prospect Theoretic Games
Abstract
Human agents in non cooperative games often exhibit non rational behaviour. Prospect theory provides a behavioural framework to capture how agents perceive gains and losses under risk, but, its incorporation into games typically leads to non concave and non convex utility functions, making equilibrium computation challenging. In such cases, it is difficult to obtain algorithms that have global convergence behaviour. This paper proposes a learning algorithm which uses minorization-maximization (MM). Under a proximal Polyak-Lojasiewicz (PL) condition on the potential function, we establish global convergence of the algorithm to a potential-maximizing Nash equilibrium. We further show how the structure of the MM minorizer can be used to enable a distributed implementation of the algorithm. With additional noise injection, the individual updates of the agents can be obscured, while still ensuring global convergence.
Introduction
Non cooperative games have been used successfully to model various problems that arise in areas such as engineering and economics. The classical framework of expected utility maximization has been used predominantly in these works. Usually, these generally assume fully rational agents. However, human agents exhibit a number of non rational behaviours, which reflect in their game play and decision making. Consequently, there have been a number of different approaches that sought to incorporate bounded rational behaviour of human agents in games. Prospect theory (PT) emerged as a framework for capturing systematic departures from expected utility maximization in human decision-making under risk.
A number of papers have used PT in game theory. However, the utility model under PT tends to be in general non-concave and non-smooth. This makes it difficult to directly apply a number of existing game theoretic tools and learning algorithms. In this paper, we consider a non cooperative game model, with agents taking actions from continuous, compact sets. The game is a potential game, and the utility is assumed to be continuously differentiable. However, no assumption is made on the concavity (or convexity) of the utility function (and consequently, the potential function). A learning approach based on minorization-maximization (MM) is proposed. It is shown that the minorization functions can be chosen such that the MM learning approach converges to a potential maximizing Nash equilibrium. The result hinges on the potential function obeying a proximal Polyak-Łojasiewicz (PŁ) inequality, which ensures that stationary points over the strategy set are maximizers of the potential.
In many applications, agents may be reluctant to share information about their personal preferences with other agents. Prospect-theoretic value functions encode such preferences by capturing how agents perceive and evaluate outcomes under risk. In this paper, we provide an MM algorithm, that can be implemented in a distributed fashion, with noise injected to limit the disclosure of their private preferences. Thus, in non-concave games, it is possible to develop learning algorithms that are distributed and still have global convergence guarantees.
Literature Review
Prospect theory [1] and its cumulative version [2] have been incorporated in non-cooperative games in several ways. Equilibrium notions for agents with PT and CPT preferences, and their existence and properties, have been studied in [3, 4, 5], which show that probability weighting can substantially change the structure of the equilibrium set. Learning in repeated games with CPT agents was considered in [6], where calibrated learning is shown to converge to a set of CPT correlated equilibria. PT has also been used in engineering applications, for instance to model prosumers in smart grid energy management [7]. These works mostly consider finite action sets or specific game structures; in contrast, we consider continuous action sets and study learning dynamics with global convergence guarantees to Nash equilibria.
In continuous games, convergence of learning dynamics is typically established under concavity or monotonicity of the utilities [8, 9]. Without such assumptions, gradient-based learning may fail to converge, or may avoid some Nash equilibria [10]. In [11], the authors propose the use of local Nash equilibria to study non-concave, non-smooth games. Potential games [12] have the useful property that learning in the game can be related to the optimization of the potential function [13]; however, when the potential is non-concave, only convergence to stationary points can be expected in general. The Polyak-Łojasiewicz (PŁ) inequality [14] provides a way of obtaining global convergence without concavity, with linear rates for gradient and proximal-gradient methods under its proximal version [15]. In games, a two-sided PŁ condition has been used to obtain global convergence of gradient methods in non convex-non concave minimax problems [16]. Here, we use the proximal PŁ inequality on the potential function of a game with PT utilities.
MM algorithms [17, 18] successively optimize surrogate functions that bound the objective and are tight at the current iterate. For non convex problems, their convergence to stationary points has been established under general conditions [19, 20]. In multi-agent problems, surrogates obtained by partially linearizing the objective lead to decompositions that can be solved in parallel by the agents [21]. Our minorizer follows a similar idea: the coupling term is kept exact, while the individual PT terms, which encode private preferences, are linearized.
Distributed algorithms for computing Nash equilibria based on the alternating direction method of multipliers (ADMM) have been proposed in [22]. When the agents’ objectives are private, a common approach is to perturb the exchanged messages with noise, which leads to a trade-off between privacy and accuracy [23]; for such message perturbation schemes, exact convergence and privacy guarantees cannot be achieved simultaneously [24]. In this paper, the noisy ADMM is analyzed as an inexact maximization step within the MM scheme, which yields convergence guarantees when the noise variance decreases.
The main contributions of this paper are as follows.
We propose an MM learning scheme for non-cooperative potential games with prospect theoretic, non concave potentials. We show that, when the potential satisfies a proximal PL inequality, the scheme converges at a linear rate to a potential maximizing Nash equilibrium.
We construct a separable minorizer that allows the maximization step to be carried out in a distributed fashion using ADMM, in which each agent’s preferences enter only its own local update.
We analyze MM with an inexact maximization step, and apply this analysis to a noisy ADMM in which agents perturb their updates to obscure their preferences. We show that the iterates converge when the noise variance decreases, up to an error that decays exponentially in the number of ADMM iterations.
Overview of Prospect Theory
We now present a high level overview of Prospect theory [1, 25]. Under the expected utility paradigm, a user evaluates their prospects based on expected utility. A prospect involves outcomes with a distribution , where the outcomes are ordered in increasing order of attractiveness. The expected utility is
Comparing to another prospect having outcomes with distribution , involves a comparison between and , and choosing the higher one.
Under PT, the perception of outcomes changes with respect to a reference point. Outcomes above this reference are seen as gains, and those below as losses. When faced with gains, agents will behave in a risk averse manner, and when faced with losses, agents tend to be risk seeking or risk neutral. This perception is captured by an outcome to value map , which, in order to model the above (and other) aspects of bounded rational behaviour, is assumed in general to be convex below a threshold, concave above it. One such function is depicted in Figure 1. Here the agents are risk averse for gains and risk seeking for losses, with the zero being the reference point that demarcates gains from losses.

Figure 1. Example of a prospect theoretic value function that maps outcomes to values.
In addition to the notion of value relative to risk, the weight accorded to outcomes is modified as well. The modified value of the prospect is
where is a function similar to the one in Figure 1, and are the weights for the distorted outcomes, given by [25],
for an appropriate weighting function , which is monotone increasing and bounded between 0 and 1. This models the overweighting of small probabilities, and the under-weighting of large probabilities, for large outcomes. Faced with a choice between two prospects and , the agent will make a decision comparing their PT modified values and , and choosing the better option. It is clear that these decisions are likely to be very different from those predicted by expected utility theory.
Game Model
We consider a game with agents, of which play the game, while agent acts as a coordinator between the agents. Let denote the set of agents that participate in the game. Agent chooses its strategy from a convex and compact set . Let denote the joint strategy vector, which is an element of . The randomness in outcomes is modelled by a real valued random variable taking values over the set where , with probability distribution . Let , and be monotone increasing. The utility for user under the strategy where , is given by
where , and is the distribution generated from by (2) using . This form of the utility function includes the following terms:
A collective usage benefit , which is a collective utility common to all agents.
An individual random reward . This reward is random due to its dependence on , and is viewed subjectively through the PT value function .
We assume that the functions , and are continuous and smooth, for each .
Definition 1. We define the game as the tuple .
Definition 2. A Nash equilibrium for is any such that for all ,
Potential Structure
The game is a weighted potential game [12], with weights and potential function
that is, for all , and . Since is continuous and is compact, the set of maxima
is non empty, and every element of is a Nash equilibrium [12]; we call such points potential optimal equilibria. Define the maximum value of the potential function and the optimality gap,
We make the following assumption.
Assumption 1 (Proximal PL Inequality). There exist and such that the potential function satisfies, for all ,
where, for ,
Assumption 1 is a generalization of the PL inequality to constrained problems [15]. The maximum in (9) is attained at , where denotes projection to the set . The function satisfies , with equality if and only if is a stationary point of over , i.e., for all . Whenever the gradient step lies in (in particular, if the constraints were absent), , and (8) reduces to the classical PL inequality . The classical PL inequality ensures that all points with zero gradient are optimal points for unconstrained optimization. The proximal PL assumption ensures that every stationary point of over , including stationary points on the boundary of , is also a maximizer. Thus, learning algorithms that move towards a stationary point are assured of reaching an optimum.
The following property of the function will be used in subsequent results.
Lemma 1. For and all ,
Proof.
The result follows by observing that any element in the set is contained in the set .
Convergence under the Proximal PŁ Condition
Definition 3 (MM Learning). Consider the sequence of joint strategies , where
where is the minorizing function at iteration , which is any function that satisfies,
.
The initial point .
For a non concave function, this MM algorithm converges to a stationary point [20]. Convergence to global maximum can be ensured under further assumptions.
Assumption 2. The minorizer is differentiable and strongly concave in , with modulus , and its gradient is -Lipschitz, uniformly over . Moreover, for every , the minorizer is tangent to the potential at , i.e., .
Theorem 2. Under Assumptions 1 and 2, the MM learning scheme converges to a potential maximizing Nash equilibrium. Moreover, with , the optimality gap
Proof. Since the minorizer is strongly concave with modulus , we have, for every ,
Since maximizes the concave function over the convex set , it satisfies the first order optimality condition for all . Using (15) with , and , we obtain
This implies,
Hence,
Following the argument of [15 Theorem 5], fix . Then,
where we used and Assumption 2. Taking and maximizing both sides over , we obtain, using the definition (9),
Using Lemma 1 and Assumption 1, we get
This yields , where since , and therefore
As , , and hence, . From (16) and using the fact that is non negative and decreasing, we can write
Thus is Cauchy, and converges to some . Since converges to , it follows that is a maximizer of the potential function .
Privacy Aware MM
While the minorizer has to be designed keeping the requirements of Theorem 2 in mind, maximization of the minorizer (which is to be done in every iteration of the MM algorithm), can itself be implemented in a partially distributed fashion. The need for a distributed implementation stems from the fact that agents’ individual reward functions encode information about their personal preferences through their prospect-theoretic value functions, which they may be reluctant to disclose to other agents.
A minorizer that enables distributed optimization at each iteration, while limiting the disclosure of agents’ preference information to one another, is obtained by keeping the collective term exact and linearizing only the individual terms. Write the potential as
with , and consider
where
If each is -Lipschitz, the descent lemma [26 Lemma 1.2.3] gives , with equality at , so that (27) is a minorizer of . Since , it is tangent to , and if is concave with -Lipschitz gradient, it satisfies Assumption 2 with and . Thus, there is a separable structure in the minorizer: the collective term couples the agents, while each depends only on the strategy and on the PT preferences of agent . If is a separable function, then the optimization becomes completely separable, and can be solved in a distributed manner. Linearizing the whole potential instead would make the MM step a projected gradient step [15], but would not preserve this separable structure.
Noisy Alternating Direction Method of Multipliers (ADMM) Method
An ADMM based approach can be used to maximize (27) if the functions and are concave. To limit the information revealed to other agents, each agent can inject noise into its individual updates, thereby making its state variables less directly observable. For a concave function , define the proximal operator,
The ADMM algorithm [27] is
The first step is generally centralized (unless is separable) and is performed by agent . The other two steps can be carried out by the agents in a distributed fashion. Importantly, the individual prospect-theoretic value functions encode agents’ personal preferences, through parameters that characterize how they evaluate outcomes under risk. Their derivatives , which enter the individual updates, may therefore reveal information about these preferences. To limit such disclosure, agents inject noise into their updates, so that the messages sent to the coordinator do not reveal the exact values of . As shown in § 5, convergence to requires the variance of this noise to decrease over the MM iterations.
Convergence with an Inexact MM Step
The noisy ADMM of § 4 can be considered an MM scheme where the optimization in the MM step returns an answer with error. We first show that MM can tolerate certain errors and then show that the noisy ADMM produces errors of the required kind, under assumptions on the collective benefit .
Throughout, , and
denotes the exact maximization step in the MM algorithm. From Assumption 2, it follows that this is unique.
MM maximization with error
The error of the inexact MM step is
Let denote the history of the algorithm up to . We also define
By the -Lipschitz gradient and the tangency of (Assumption 2), , so that . Moreover, when lies in the interior of for every , since at an interior maximizer.
Theorem 3 (MM with error). Let Assumptions 1 and 2 hold, let , and suppose that for a deterministic sequence . Then
and, with ,
In particular, if .
Proof. Fix , and write and , so that . By the minorization property and the -Lipschitz gradient of (Assumption 2),
As in (20), by Lemma 1 and Assumption 1, , and subtracting both sides from gives
Taking conditional expectations, with by Jensen’s inequality, and unrolling the recursion gives (36). Finally, the weights in (36) sum to at most , which gives (37).
Theorem 3 makes no assumption on how the MM step is solved: the errors need not be zero-mean or independent, and only their conditional second moments are used. The errors enter (36) through two terms. The linear term is present only when lies on the boundary of . When is interior for all , and the errors have a quadratic effect, so that the asymptotic gap scales with the variance of the error. Finally, since is continuous on the compact set , implies that the distance from to the set of maximizers tends to zero; when has a unique maximizer , .
Noisy ADMM with finite iterations
We now consider the case where the MM step is solved by the noisy ADMM of § 4. At MM step , with , the algorithm runs iterations of (30)–(32), with , from and , and sets , the projection of the last iterate onto . With the state , the noiseless iteration () is a map , and the noisy one reads
since the noise added to is subtracted in the update of .
Lemma 4 (Error of noisy ADMM). Let the minorizer be (27), with and for some and , and suppose that, for every , the maximizer of over lies in . If for , then there are constants and , independent of , such that
Proof. Since and are quadratic, the noiseless iteration is affine, : the linear parts of the proximal operators, and , do not depend on , while depends continuously on through and . As the noiseless ADMM converges linearly to the maximizer of [28, 29], for some , and has a unique fixed point , with . Unrolling (39),
Since the proximal operator and the projection onto are nonexpansive, . By Minkowski’s inequality, using ,
which gives (40), since is bounded over the compact set .
After inner iterations, the initial error has shrunk as , while the accumulated noise remains of order . The interior assumption is needed because the ADMM of § 4 does not take the constraints into account. Otherwise, each agent can project its update onto , and a similar bound follows from the contraction property of Douglas–Rachford splitting [28].
Corollary 5 (MM with noisy ADMM). Under the assumptions of Theorem 3 and Lemma 4, the bound (36) holds with given by (40). In particular, if ,
where .
Proof. By Lemma 4, Theorem 3 applies with given by (40). If , then , and (37) gives (41).
The bound of Corollary 5 separates three effects: the initial gap is forgotten at the rate of Theorem 2; the noise contributes through and vanishes only if its variance decreases to zero, independent and identically distributed noise leaving an asymptotic gap; and the finite number of inner iterations leaves the floor (41), which decreases exponentially in and vanishes if grows with .
The injected noise only obscures the agents’ individual updates. Since its variance must decrease for the iterates to converge, the level of obfuscation it provides also diminishes over time. This highlights a trade-off between convergence accuracy and privacy protection in noise-based mechanisms: for schemes that perturb the exchanged messages, exact convergence and differential privacy cannot be achieved simultaneously [24]. Larger noise levels may provide stronger obfuscation of individual preference information, but can also increase the error in the MM updates and slow down convergence. Conversely, reducing the noise improves solution accuracy and convergence, but progressively weakens the protection of individual information.
Numerical Example
We illustrate our results on an example which captures the key features of our analysis: agents with smooth S-shaped value functions that make the potential non-concave while satisfying our assumptions; a collective benefit that couples the players; a game small enough for the optimum to be computed exactly. We consider agents and a coordinator (agent ). Agent chooses , interpreted as a quantity sold at a random price with probabilities . The reward of agent is , where is a reference level. Outcomes are evaluated through the smooth prospect-theoretic value function
which is increasing, concave on gains and convex on losses (Figure 2).
Figure 2. Prospect-theoretic value functions used in the example: concave on gains () with saturation level , convex on losses () with saturation level , where .
The weights are obtained from via (2) with , , giving . The collective benefit is
Its first part penalizes, quadratically, the deviation of each agent from its share of a collective target (as in tracking or imbalance penalties), while the second part penalizes disparities between agents and thus favours an equitable distribution of effort. It can also be written as
with , where is the Laplacian of the complete graph. This form is also chosen for tractability of the coordinator’s proximal step in the ADMM. The utilities are . The different parameters are listed in Table 1.
| Agent 1 | Agent 2 | Agent 3 | |
| 1.0 | 0.8 | 0.5 | |
| 2.0 | 1.5 | 2.5 | |
| 1.0 | 1.5 | 0.8 | |
| 1.5 | 2.0 | 1.2 | |
Table 1. Parameters of the numerical example.
When , one can show that the potential has a unique critical point, which is its global maximizer , lying in the interior of , and that satisfies Assumption 1 (numerically, for any ). Nevertheless, and each are non-concave: near .
Algorithm settings At each MM step the minorizer (27) of § 4, , with
and , is maximized by 30 iterations of noisy ADMM with . Note that is a minorizer of whenever for all . We take equal to the closed-form bound obtained from with ,
which guarantees minorization and can be computed without any numerical search. At MM step , agents add independent noise with , for initial variances . Each configuration is run 20 times from for 150 MM steps, and the reported curves are averages over runs.
Results Figure 3 shows the potential along the MM iterates, and Figure 4 the optimality gap on a logarithmic scale. For every noise level the iterates converge to the potential maximizing equilibrium . A larger initial variance slows the early iterations, where the noise is largest, but since it does not leave a residual gap: after 150 MM steps the mean gap ranges from without noise to for . With noise, the gap decays geometrically at a rate governed by the noise decay rather than by the MM contraction alone. This is the behaviour predicted by Corollary 5: in this example the exact MM step is interior, so that , and since the noise variance decays to zero, the only remaining floor is due to the finite number of ADMM iterations, which is negligible for .

Figure 3. Potential along the MM iterates for different initial noise variances, averaged over 20 runs; shaded regions show the 10–90% range across runs. The dashed line is .

Figure 4. Optimality gap (log scale) for different initial noise variances, averaged over 20 runs; shaded regions show the 10–90% range across runs.
Impact of prospect theory If we consider the same game with expected-utility players, i.e., and , the potential is a concave quadratic, and the equilibrium is , roughly twice the PT equilibrium . The difference is driven by the value function: at equilibrium all outcomes are gains, where is concave and saturates, so the PT players, being risk averse in gains, sell much smaller quantities. Probability weighting alone has the opposite effect: using with gives . This is because the weighting overweights the rare high price ( against ), which raises the perceived expected price ( against ). Also note that PT is also what makes the problem non-concave: the non-concavity of comes from the convex loss region of , and is absent under expected utility, where standard methods apply.
Effect of the minorizer curvature The bound used above is conservative. For the minorizer (27), the condition of § 4 ( -Lipschitz) holds for , which here coincides with , the smallest for which is a minorizer; for the minorization is imperfect. To study the effect of , we run the full algorithm for a fixed budget of MM steps, for 21 values of and the same decaying noise as above (20 runs per configuration, 150 ADMM iterations per MM step). Figure 5 shows the resulting value , with the dotted line marking .

Figure 5. Potential after MM steps versus the curvature parameter , for different initial noise variances, averaged over 20 runs; shaded regions show the 10–90% range across runs. The dotted line marks , and the dashed line is . The horizontal axis is logarithmic in plus a small constant (0.2), so that can be shown.
To the right of , a larger makes more curved and each MM step shorter, so after steps the iterates are further from the optimum: at the mean of is about 2.3 without noise and about 0 for , against , and the noise also widens the spread across runs. To the left of , where is no longer a minorizer and Theorem 2 does not apply, the curves stay close to : in all runs the algorithm still converged to . In this region the remaining gap is set by the noise rather than by ; it is below without noise and at most , and 0.13 for , 1 and 5. Choosing above thus slows convergence, while values below can be faster but are not covered by Theorem 2.
Conclusion
We studied learning in non-cooperative potential games with prospect theoretic non concave utilities. If the potential satisfies a proximal PL inequality, an MM learning algorithm can be constructed, which converges linearly to a potential maximizing Nash equilibrium. The minorizer for this algorithm can be designed to enable distributed implementation using a noisy ADMM. The noise prevents the agents from revealing their preferences exactly. Convergence of the noisy MM algorithm is shown under assumptions of the noise variance. In particular, we demonstrate a trade off between algorithm convergence and privacy for agent preferences: noise levels that do not decay in time ensure better hiding of agent preferences, whereas decaying noise levels lead to convergence to equilibrium. Numerical results illustrate the performance of the proposed algorithm. A formal characterization of the privacy guarantee will be undertaken in a follow-up work.
Declaration of Generative AI and AI-Assisted Technologies
The authors used Claude (Opus 5.5, Anthropic) to assist in checking and deriving intermediate algebraic steps and formatting mathematical derivations in Sections 4 and 6, for designing the numerical simulations in Section 7, and for formatting and editing the paper. The authors have independently re-derived and verified all the content of the paper, and take full responsibility for the mathematical accuracy and content of the publication.
References
- [1]D. Kahneman and A. Tversky, “Prospect theory: An analysis of decision under risk,” Econometrica, vol. 47, no. 2, pp. 363–391, 1979.
- [2]A. Tversky and D. Kahneman, “Advances in prospect theory: Cumulative representation of uncertainty,” Journal of Risk and Uncertainty, vol. 5, no. 4, pp. 297–323, 1992.
- [3]K. Keskin, “Equilibrium notions for agents with cumulative prospect theory preferences,” Decision Analysis, vol. 13, no. 3, pp. 192–208, 2016.DOI
- [4]L. P. Metzger and M. O. Rieger, “Non-cooperative games with prospect theory players and dominated strategies,” Games and Economic Behavior, vol. 115, pp. 396–409, 2019.DOI
- [5]S. R. Phade and V. Anantharam, “On the geometry of Nash and correlated equilibria with cumulative prospect theoretic preferences,” Decision Analysis, vol. 16, no. 2, pp. 142–156, 2019.
- [6]S. R. Phade and V. Anantharam, “Learning in games with cumulative prospect theoretic preferences,” Dynamic Games and Applications, vol. 13, no. 1, pp. 265–306, 2023.
- [7]S. R. Etesami, W. Saad, N. B. Mandayam, and H. V. Poor, “Stochastic games for the smart grid energy management with prospect prosumers,” IEEE Transactions on Automatic Control, vol. 63, no. 8, pp. 2327–2342, 2018.
- [8]A. Krishnan K.S., H. Le Cadre, and A. Bušić, “How irrationality shapes Nash equilibria: A prospect-theoretic perspective,” in 2025 IEEE 64th Conference on Decision and Control (CDC), pp. 4428–4433, IEEE, 2025.
- [9]A. Krishnan K.S., H. Le Cadre, and A. Bušić, “Achieving a Collective Target through Incentives,” in International Conference on Network Games, Artificial Intelligence, Control and Optimization, pp. 57–67, Springer, 2025.
- [10]E. Mazumdar, L. J. Ratliff, and S. S. Sastry, “On gradient-based learning in continuous games,” SIAM Journal on Mathematics of Data Science, vol. 2, no. 1, pp. 103–131, 2020.DOI
- [11]M. Fochesato, F. Pokou, H. Le Cadre, and J. Lygeros, “Noncooperative games with prospect theoretic preferences,” IEEE Control Systems Letters, 2025.DOI
- [12]D. Monderer and L. S. Shapley, “Potential games,” Games and economic behavior, vol. 14, no. 1, pp. 124–143, 1996.
- [13]J. R. Mardeen, G. Arslan, and J. S. Shamma, “Cooperative control and potential games,” IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics), vol. 39, no. 6, pp. 1393–1407, 2009.
- [14]B. T. Polyak, “Gradient methods for minimizing functionals,” USSR Computational Mathematics and Mathematical Physics, vol. 3, no. 4, pp. 864–878, 1963.
- [15]H. Karimi, J. Nutini, and M. Schmidt, “Linear convergence of gradient and proximal-gradient methods under the Polyak-Lojasiewicz condition.” arXiv:1608.04636, 2016.arxiv.org/abs/1608.04636
- [16]J. Yang, N. Kiyavash, and N. He, “Global convergence and variance reduction for a class of nonconvex-nonconcave minimax problems,” in Advances in Neural Information Processing Systems, vol. 33, 2020.
- [17]D. R. Hunter and K. Lange, “A tutorial on MM algorithms,” The American Statistician, vol. 58, no. 1, pp. 30–37, 2004.DOI
- [18]Y. Sun, P. Babu, and D. P. Palomar, “Majorization-minimization algorithms in signal processing, communications, and machine learning,” IEEE Transactions on Signal Processing, vol. 65, no. 3, pp. 794–816, 2017.DOI
- [19]M. Razaviyayn, M. Hong, and Z.-Q. Luo, “A unified convergence analysis of block successive minimization methods for nonsmooth optimization,” SIAM Journal on Optimization, vol. 23, no. 2, pp. 1126–1153, 2013.arxiv.org/abs/1209.2385
- [20]J. Bolte and E. Pauwels, “Majorization-minimization procedures and convergence of sqp methods for semi-algebraic and tame programs,” Mathematics of Operations Research, vol. 41, no. 2, pp. 442–465, 2016.DOI
- [21]G. Scutari, F. Facchinei, P. Song, D. P. Palomar, and J.-S. Pang, “Decomposition by partial linearization: Parallel optimization of multi-agent systems,” IEEE Transactions on Signal Processing, vol. 62, no. 3, pp. 641–656, 2014.DOI
- [22]F. Salehisadaghiani, W. Shi, and L. Pavel, “Distributed Nash equilibrium seeking under partial-decision information via the alternating direction method of multipliers,” Automatica, vol. 103, pp. 27–35, 2019.DOI
- [23]Z. Huang, S. Mitra, and N. Vaidya, “Differentially private distributed optimization,” in Proceedings of the International Conference on Distributed Computing and Networking (ICDCN), pp. 4:1–4:10, ACM, 2015.arxiv.org/abs/1401.2596
- [24]E. Nozari, P. Tallapragada, and J. Cortés, “Differentially private distributed convex optimization via functional perturbation,” IEEE Transactions on Control of Network Systems, vol. 5, no. 1, pp. 395–408, 2018.
- [25]H. P. Stott, “Cumulative prospect theory’s functional menagerie,” Journal of Risk and uncertainty, vol. 32, pp. 101–130, 2006.DOI
- [26]Y. Nesterov, Introductory Lectures on Convex Optimization: A Basic Course, vol. 87 of Applied Optimization. Boston, MA: Kluwer Academic Publishers, 2004.
- [27]N. Parikh and S. Boyd, “Proximal algorithms,” Foundations and Trends in optimization, vol. 1, no. 3, pp. 127–239, 2014.DOI
- [28]P. Giselsson and S. Boyd, “Linear convergence and metric selection for douglas-rachford splitting and admm,” IEEE Transactions on Automatic Control, vol. 62, no. 2, pp. 532–544, 2016.arxiv.org/abs/1410.8479
- [29]W. Deng and W. Yin, “On the global and linear convergence of the generalized alternating direction method of multipliers,” Journal of Scientific Computing, vol. 66, no. 3, pp. 889–916, 2016.
