arXiv research

Optimal strategies are found for a repeated betting game using diffusion approximation.

Study explores algorithmic collusion in repeated games using various learning dynamics.

LAFF algorithm balances adaptability and non-exploitability in repeated games.

This paper examines transitions in sniping behavior among algorithmic traders, finding new profitable strategies.

New method detects heuristics in complex game strategies.

Algorithm learns to play against unknown opponents in sequential games.

R2-B2 optimizes game interactions with recursive reasoning.

No-regret learning fails to converge to Nash equilibria in mixed strategies.

We consider the dynamics of player's strategies in repeated market games, where the selection of strategies is determined by a learning model. Prior theoretical analysis and experimental data show that after large number of plays the average number of agents who decide to enter, per round of the game, approaches the ma…

We describe an approximate dynamic programming (ADP) approach to compute approximations of the optimal strategies and of the minimal losses that can be guaranteed in discounted repeated games with vector-valued losses. Such games prominently arise in the analysis of regret in repeated decision-making in adversarial env…

A study on how a principal can incentivize an agent to make better decisions in a repeated game.

We consider regret minimization in repeated games with non-convex loss functions. Minimizing the standard notion of regret is computationally intractable. Thus, we define a natural notion of regret which permits efficient optimization and generalizes offline guarantees for convergence to an approximate local optimum. W…

We consider Blackwell approachability, a very powerful and geometric tool in game theory, used for example to design strategies of the uninformed player in repeated games with incomplete information. We extend this theory to "generalized quitting games" , a class of repeated stochastic games in which each player may ha…

Algorithm improves RL model selection for repeated games with utility maximization.

Game theory helps machine learn better from adversarial queries.

Study of repeated games with unobserved agent rewards using MAB framework.

New algorithm reduces online learning regret for bounded recall games.

Playing repeated matrix games (RMG) while maximizing the cumulative returns is a basic method to evaluate multi-agent learning (MAL) algorithms. Previous work has shown that $UCB$, $M3$, $S$ or $Exp3$ algorithms have good behaviours on average in RMG. Besides, hedging algorithms have been shown to be effective on predi…

Study shows how adaptive market agents can lead to persistent overpricing in financial markets.

The paper analyzes how mutable blockchain protocols affect miner behavior and strategic stability.

Human behavioural patterns exhibit selfish or competitive, as well as selfless or altruistic tendencies, both of which have demonstrable effects on human social and economic activity. In behavioural economics, such effects have traditionally been illustrated experimentally via simple games like the dictator and ultimat…

Algorithm learns from changing zero-sum games with no regret.

Study on HFTs' interactions with a large trader using mean field game theory.

New algorithm reduces risk in online games with limited feedback.

New approach for adaptive conformal inference using Blackwell's theory.

Study shows market makers can cooperate without communication.

NeuPL learns diverse policies in strategy games efficiently.

New algorithms minimize regret with global costs in online learning.

New algorithms converge faster to Nash equilibrium in zero-sum games with bandit feedback.

We study the problem of repeated play in a zero-sum game in which the payoff matrix may change, in a possibly adversarial fashion, on each round; we call these Online Matrix Games. Finding the Nash Equilibrium (NE) of a two player zero-sum game is core to many problems in statistics, optimization, and economics, and fo…

Study of repeated principal-agent bandit game with self-interested and exploratory learning agents.

We consider the problem of learning to play a repeated multi-agent game with an unknown reward function. Single player online learning algorithms attain strong regret bounds when provided with full information feedback, which unfortunately is unavailable in many real-world scenarios. Bandit feedback alone, i.e., observ…

User strategization undermines algorithmic trustworthiness.

Extends trading framework to incorporate real-world constraints.

Generative adversarial networks (GANs) are successful deep generative models. GANs are based on a two-player minimax game. However, the objective function derived in the original motivation is changed to obtain stronger gradients when learning the generator. We propose a novel algorithm that repeats the density ratio e…

We consider the problem of influence maximization in fixed networks for contagion models in an adversarial setting. The goal is to select an optimal set of nodes to seed the influence process, such that the number of influenced nodes at the conclusion of the campaign is as large as possible. We formulate the problem as…

We introduce a criterion how to price derivatives in incomplete markets, based on the theory of growth optimal strategy in repeated multiplicative games. We present reasons why these growth-optimal strategies should be particularly relevant to the problem of pricing derivatives. We compare our result with other alterna…

We introduce and discuss a general criterion for the derivative pricing in the general situation of incomplete markets, we refer to it as the No Almost Sure Arbitrage Principle. This approach is based on the theory of optimal strategy in repeated multiplicative games originally introduced by Kelly. As particular cases …

The notion of \emph{policy regret} in online learning is a well defined? performance measure for the common scenario of adaptive adversaries, which more traditional quantities such as external regret do not take into account. We revisit the notion of policy regret and first show that there are online learning settings …

Fictitious play is a simple and widely studied adaptive heuristic for playing repeated games. It is well known that fictitious play fails to be Hannan consistent. Several variants of fictitious play including regret matching, generalized regret matching and smooth fictitious play, are known to be Hannan consistent. In …

First sample-efficient algorithm for learning EFCE in bandit feedback settings.

We consider a non-stochastic online learning approach to price financial options by modeling the market dynamic as a repeated game between the nature (adversary) and the investor. We demonstrate that such framework yields analogous structure as the Black-Scholes model, the widely popular option pricing model in stochas…

We consider Bandits with Knapsacks (henceforth, BwK), a general model for multi-armed bandits under supply/budget constraints. In particular, a bandit algorithm needs to solve a well-known knapsack problem: find an optimal packing of items into a limited-size knapsack. The BwK problem is a common generalization of nume…

Learning by experience in Multi-Agent Systems (MAS) is a difficult and exciting task, due to the lack of stationarity of the environment, whose dynamics evolves as the population learns. In order to design scalable algorithms for systems with a large population of interacting agents (e.g. swarms), this paper focuses on…

Maker-taker fees can prevent algorithmic cooperation in market making, but not always.

Reinforcement learning aids decision-making in economics and finance.

This paper explores using nonlinear control for robust logarithmic growth in coin flipping games.

Can machine learning models for recommendation be easily fooled? While the question has been answered for hand-engineered fake user profiles, it has not been explored for machine learned adversarial attacks. This paper attempts to close this gap. We propose a framework for generating fake user profiles which, when inco…