Even with an unprecedented breakthrough of deep learning in electroencephalography (EEG), collecting adequate labelled samples is a critical problem due to laborious and time‐consuming labelling. Recent study proposed to solve the limited label problem via domain adaptation methods. However, they mainly focus on reducing domain discrepancy without considering task‐specific decision boundaries, which may lead to feature distribution overmatching and therefore make it hard to match within a large domain gap completely. A novel self‐training maximum classifier discrepancy method for EEG classification is proposed in this study. The proposed approach detects samples from a new subject beyond the support of the existing source subjects by maximising the discrepancies between two classifiers' outputs. Besides, a self‐training method that uses unlabelled test data to fully use knowledge from the new subject and further reduce the domain gap is proposed. Finally, a 3D Cube that incorporates the spatial and frequency information of the EEG data to create input features of a Convolutional Neural Network (CNN) is constructed. Extensive experiments on SEED and SEED‐IV are conducted. The experimental evaluations exhibit that the proposed method can effectively deal with domain transfer problems and achieve better performance.

Modelling the learning dynamic of multi-agent systems has long been a crucial issue for understanding the emergence of collective behavior. In public goods games, agents interact in multiple larger groups. While previous studies have primarily focused on infinite populations that only allow pairwise interactions, we aim to investigate the learning dynamics of agents in a public goods game with higher-order interactions. With a novel use of hypergraphs for encoding higher-order interactions, we develop a formal model (a Fokker-Planck equation) to describe the temporal evolution of the distribution function of Qvalues. Noting that early research focused on replicator models to predict system dynamics failed to accurately capture the impact of hyperdegree in hypergraphs, our model effectively maps its influence. Through experiments, we demonstrate that our theoretical findings are consistent with the agent-based simulation results. We demonstrated that as the number of groups an agent is involved in reaches a certain scale, the learning dynamics of the system evolve to resemble those of a well-mixed population. Furthermore, we demonstrate that our model offers insights into algorithmic parameters, such as the Boltzmann temperature, facilitating parameter tuning

Coevolution is considered as an effective means to optimize the conditions for the survival of cooperation. In this work, we propose a coevolution rule between individuals’ node weights and aspiration, and then explore how this mechanism affects the evolution of cooperation in the spatial prisoner’s dilemma game. We show that there is an optimistic amplitude of node weights that guarantees the survival of cooperation even when temptation to antisocial behavior is relatively large. An explanation is provided from a microscopic point of view by dividing nodes into four different types. What is interesting, our coevolution rule results in spontaneous emergence of cyclic dominance, where defectors with low weight become cooperators by imitating cooperators with high weight.

In this article, we introduce a new theory for payoff control in multichannel learning environments, where agents interact with each other over multiple channels and each channel is a repeated normal form game. We propose two payoff control strategies—partial control and full control—that allow a single agent to set an upper bound to the opponent’s expected payoffs summed across all channels, even if the opponent is a reinforcement learning agent. We prove that a partial (or full) control strategy can be obtained by solving a system of inequalities, and characterize the conditions under which such a partial (or full) control strategy exists. We show that by utilizing these control strategies, the agent can influence the opponent’s learning evolution and direct it toward a desired viable equilibrium. Our experiments confirm the effectiveness of our theory for payoff control in a wide range of multichannel learning environments.

Stochastic games form the foundational mathematical framework for describing multiagent interactions and underpin the theoretical foundations of multiagent reinforcement learning (MARL) and optimal decision making. However, previous research has typically focused on either two-agent settings or large-scale well-mixed agent populations, where the considered interaction scenarios were far from realistic. In this article, we consider structured populations where agents can interact with immediate neighbors. By using the pair-approximation method, we develop a new dynamical model to describe the $Q$ -learning dynamics in stochastic games on regular graphs. Through comparisons with agent-based simulation results, we validate the accuracy of our dynamical model across various stochastic games, population structures, and algorithm parameters. Our research thus provides both qualitative and quantitative insights into the effects of state transition rules and graph topologies in population dynamics. In particular, we show that, under certain conditions, state transitions can significantly promote the evolution of cooperation in social dilemmas. We also explored the effects of agent degree on cooperation, and unlike previous findings, we show that this can have either positive or negative implications for cooperation depending on the transition rules.