Tom Lenaerts

JAIR Journal 2025 Journal Article

Collective Intelligence in Decision-Making with Non-Stationary Experts

• Axel Abels
• Vito Trianni
• Ann Nowé
• Tom Lenaerts

When sufficient experience to make informed decisions is unavailable, expert advice can help us navigate uncertainty. As expertise evolves, driven by continuous learning in human experts or model updates in artificial experts, it is crucial to adopt adaptive approaches. Existing methods for exploiting non-stationary experts focus on competing with the single best expert. In contrast, this work harnesses the power of collective intelligence to facilitate better decision-making in the face of evolving expertise or dynamic environments. To achieve this, we propose the novel CORVAL approach which optimally combines the insights of multiple experts. By adapting to drifts in expertise, our novel approach can surpass the performance of the single best expert as well as previous approaches. Empirical evaluations on a diverse range of non-stationary problems, including active learning applications, showcase the improved performance of our approach in collective decision-making scenarios.

IJCAI Conference 2025 Conference Paper

Wisdom from Diversity: Bias Mitigation Through Hybrid Human-LLM Crowds

• Axel Abels
• Tom Lenaerts

Despite their performance, large language models (LLMs) can inadvertently perpetuate biases found in the data they are trained on. By analyzing LLM responses to bias-eliciting headlines, we find that these models often mirror human biases. To address this, we explore crowd-based strategies for mitigating bias through response aggregation. We first demonstrate that simply averaging responses from multiple LLMs, intended to leverage the ``wisdom of the crowd", can exacerbate existing biases due to the limited diversity within LLM crowds. In contrast, we show that locally weighted aggregation methods more effectively leverage the wisdom of the LLM crowd, achieving both bias mitigation and improved accuracy. Finally, recognizing the complementary strengths of LLMs (accuracy) and humans (diversity), we demonstrate that hybrid crowds containing both significantly enhance performance and further reduce biases across ethnic and gender-related contexts.

IJCAI Conference 2024 Conference Paper

To Promote Full Cooperation in Social Dilemmas, Agents Need to Unlearn Loyalty

• Chin-wing Leung
• Tom Lenaerts
• Paolo Turrini

If given the choice, what strategy should agents use to switch partners in strategic social interactions? While many analyses have been performed on specific switching heuristics, showing how and when these lead to more cooperation, no insights have been provided into which rule will actually be learnt by agents when given the freedom to do so. Starting from a baseline model that has demonstrated the potential of rewiring for cooperation, we provide answers to this question over the full spectrum of social dilemmas. Multi-agent Q-learning with Boltzmann exploration is used to learn when to sever or maintain an association. In both the Prisoner's Dilemma and the Stag Hunt games we observe that the Out-for-Tat rewiring rule, breaking ties with other agents choosing socially undesirable actions, becomes dominant, confirming at the same time that cooperation flourishes when rewiring is fast enough relative to imitation. Nonetheless, in the transitory region before full cooperation, a Stay strategy, keeping a connection at all costs, remains present, which shows that loyalty needs to be overcome for full cooperation to emerge. In conclusion, individuals learn cooperation-promoting rewiring rules but need to overcome a kind of loyalty to achieve full cooperation in the full spectrum of social dilemmas.

AIJ Journal 2023 Journal Article

Dealing with expert bias in collective decision-making

• Axel Abels
• Tom Lenaerts
• Vito Trianni
• Ann Nowé

Quite some real-world problems can be formulated as decision-making problems wherein one must repeatedly make an appropriate choice from a set of alternatives. Multiple expert judgments, whether human or artificial, can help in taking correct decisions, especially when exploration of alternative solutions is costly. As expert opinions might deviate, the problem of finding the right alternative can be approached as a collective decision making problem (CDM) via aggregation of independent judgments. Current state-of-the-art approaches focus on efficiently finding the optimal expert, and thus perform poorly if all experts are not qualified or if they display consistent biases, thereby potentially derailing the decision-making process. In this paper, we propose a new algorithmic approach based on contextual multi-armed bandit problems (CMAB) to identify and counteract such biased expertise. We explore homogeneous, heterogeneous and polarized expert groups and show that this approach is able to effectively exploit the collective expertise, outperforming state-of-the-art methods, especially when the quality of the provided expertise degrades. Our novel CMAB-inspired approach achieves a higher final performance and does so while converging more rapidly than previous adaptive algorithms.

ICML Conference 2023 Conference Paper

Expertise Trees Resolve Knowledge Limitations in Collective Decision-Making

• Axel Abels
• Tom Lenaerts
• Vito Trianni
• Ann Nowé

Experts advising decision-makers are likely to display expertise which varies as a function of the problem instance. In practice, this may lead to sub-optimal or discriminatory decisions against minority cases. In this work, we model such changes in depth and breadth of knowledge as a partitioning of the problem space into regions of differing expertise. We provide here new algorithms that explicitly consider and adapt to the relationship between problem instances and experts’ knowledge. We first propose and highlight the drawbacks of a naive approach based on nearest neighbor queries. To address these drawbacks we then introduce a novel algorithm — expertise trees — that constructs decision trees enabling the learner to select appropriate models. We provide theoretical insights and empirically validate the improved performance of our novel approach on a range of problems for which existing methods proved to be inadequate.

EWRL Workshop 2023 Workshop Paper

Laser Learning Environment: Insights on the challenges of coordination-critical multi-agent tasks

• Yannick Molinghen
• Raphaël Avalos
• Mark Van Achter
• Ann Nowe
• Tom Lenaerts

We introduce the Laser Learning Environment (LLE), a collaborative multi-agent reinforcement learning environment in which coordination is central. In LLE, agents depend on each other to make progress (interdependence), must jointly take specific sequences of actions to succeed (perfect coordination), and accomplishing those joint actions does not yield any intermediate reward (zero-incentive dynamics). The challenge of such problems lies in the difficulty of escaping state space bottlenecks caused by interdependence steps since escaping those bottlenecks is not rewarded. We test multiple state-of-the-art value-based MARL algorithms against LLE and show that they consistently fail at the collaborative task because of their inability to escape state space bottlenecks, even though they successfully achieve perfect coordination. We show that Q-learning extensions such as prioritised experience replay and n-steps return hinder exploration in environments with zero-incentive dynamics, and find that intrinsic curiosity with random network distillation is not sufficient to escape those bottlenecks. We demonstrate the need for novel methods to solve this problem and the relevance of LLE as cooperative MARL benchmark.

JAIR Journal 2020 Journal Article

To Regulate or Not: A Social Dynamics Analysis of an Idealised AI Race

• The Anh Han
• Luis Moniz Pereira
• Francisco C. Santos
• Tom Lenaerts

Rapid technological advancements in Artificial Intelligence (AI), as well as the growing deployment of intelligent technologies in new application domains, have generated serious anxiety and a fear of missing out among different stake-holders, fostering a racing narrative. Whether real or not, the belief in such a race for domain supremacy through AI, can make it real simply from its consequences, as put forward by the Thomas theorem. These consequences may be negative, as racing for technological supremacy creates a complex ecology of choices that could push stake-holders to underestimate or even ignore ethical and safety procedures. As a consequence, different actors are urging to consider both the normative and social impact of these technological advancements, contemplating the use of the precautionary principle in AI innovation and research. Yet, given the breadth and depth of AI and its advances, it is difficult to assess which technology needs regulation and when. As there is no easy access to data describing this alleged AI race, theoretical models are necessary to understand its potential dynamics, allowing for the identification of when procedures need to be put in place to favour outcomes beneficial for all. We show that, next to the risks of setbacks and being reprimanded for unsafe behaviour, the time-scale in which domain supremacy can be achieved plays a crucial role. When this can be achieved in a short term, those who completely ignore the safety precautions are bound to win the race but at a cost to society, apparently requiring regulatory actions. Our analysis reveals that imposing regulations for all risk and timing conditions may not have the anticipated effect as only for specific conditions a dilemma arises between what is individually preferred and globally beneficial. Similar observations can be made for the long-term development case. Yet different from the short-term situation, conditions can be identified that require the promotion of risk-taking as opposed to compliance with safety regulations in order to improve social welfare. These results remain robust both when two or several actors are involved in the race and when collective rather than individual setbacks are produced by risk-taking behaviour. When defining codes of conduct and regulatory policies for applications of AI, a clear understanding of the time-scale of the race is thus required, as this may induce important non-trivial effects. This article is part of the special track on AI and Society.

ICML Conference 2019 Conference Paper

Dynamic Weights in Multi-Objective Deep Reinforcement Learning

• Axel Abels
• Diederik M. Roijers
• Tom Lenaerts
• Ann Nowé
• Denis Steckelmacher

Many real-world decision problems are characterized by multiple conflicting objectives which must be balanced based on their relative importance. In the dynamic weights setting the relative importance changes over time and specialized algorithms that deal with such change, such as a tabular Reinforcement Learning (RL) algorithm by Natarajan and Tadepalli (2005), are required. However, this earlier work is not feasible for RL settings that necessitate the use of function approximators. We generalize across weight changes and high-dimensional inputs by proposing a multi-objective Q-network whose outputs are conditioned on the relative importance of objectives and we introduce Diverse Experience Replay (DER) to counter the inherent non-stationarity of the Dynamic Weights setting. We perform an extensive experimental evaluation and compare our methods to adapted algorithms from Deep Multi-Task/Multi-Objective Reinforcement Learning and show that our proposed network in combination with DER dominates these adapted algorithms across weight change scenarios and problem domains.

AIIM Journal 2019 Journal Article

Using game theory and decision decomposition to effectively discern and characterise bi-locus diseases

• Nassim Versbraegen
• Aziz Fouché
• Charlotte Nachtegael
• Sofia Papadimitriou
• Andrea Gazzo
• Guillaume Smits
• Tom Lenaerts

In order to gain insight into oligogenic disorders, understanding those involving bi-locus variant combinations appears to be key. In prior work, we showed that features at multiple biological scales can already be used to discriminate among two types, i. e. disorders involving true digenic and modifier combinations. The current study expands this machine learning work towards dual molecular diagnosis cases, providing a classifier able to effectively distinguish between these three types. To reach this goal and gain an in-depth understanding of the decision process, game theory and tree decomposition techniques are applied to random forest predictors to investigate the relevance of feature combinations in the prediction. A machine learning model with high discrimination capabilities was developed, effectively differentiating the three classes in a biologically meaningful manner. Combining prediction interpretation and statistical analysis, we propose a biologically meaningful characterization of each class relying on specific feature strengths. Figuring out how biological characteristics shift samples towards one of three classes provides clinically relevant insight into the underlying biological processes as well as the disease itself.

AAAI Conference 2017 Conference Paper

Centralized versus Personalized Commitments and Their Influence on Cooperation in Group Interactions

• The Anh Han
• Luis Moniz Pereira
• Luis A. Martinez-Vaquero
• Tom Lenaerts

Before engaging in a group venture agents may seek commitments from other members in the group and, based on the level of participation (i. e. the number of actually committed participants), decide whether it is worth joining the venture. Alternatively, agents can delegate this costly process to a (beneficent or noncostly) third-party, who helps seek commitments from the agents. Using methods from Evolutionary Game Theory, this paper shows that, in the context of Public Goods Game, much higher levels of cooperation can be achieved through such centralized commitment management. It provides a more efficient mechanism for dealing with commitment free-riders, those who are not willing to bear the cost of arranging commitments whilst enjoying the benefits provided by the paying commitment proposers. We show also that the participation level plays a crucial role in the decision of whether an agreement should be formed; namely, it needs to be more strict in the centralized system for the agreement to be formed; however, once it is done right, it is much more beneficial in terms of the level of cooperation as well as the attainable social welfare. In short, our analysis provides important insights for the design of multi-agent systems that rely on commitments to monitor agents’ cooperative behavior.

AAMAS Conference 2017 Conference Paper

Commitment and Participation in Public Goods Games (JAAMAS Extended Abstract)

• The Anh Han
• Luis Moniz Pereira
• Tom Lenaerts

AAAI Conference 2017 Short Paper

Coordinating Human and Agent Behavior in Collective-Risk Scenarios

• Elias Fern‡ndez Domingos
• Juan Burguillo
• Ann NowŽ
• Tom Lenaerts

Various social situations entail a collective risk. A well-known example is climate change, wherein the risk of a future environmental disaster clashes with the immediate economic interest of developed and developing countries. The collective-risk game operationalizes this kind of situations. The decision process of the participants is determined by how good they are in evaluating the probability of future risk as well as their ability to anticipate the actions of the opponents. Anticipatory behavior contrasts with the reactive theories often used to analyze social dilemmas. Our initial work can already show that anticipative agents are a better model to human behavior than reactive ones. All the agents we studied used a recurrent neural network, however, only the ones that used it to predict future outcomes (anticipative agents) were able to account for changes in the context of games, a behavior also observed in experiments with humans. This extended abstract aims to explain how we wish to investigate anticipation within the context of the collective-risk game and the relevance these results may have for the field of hybrid socio-technical systems.

AAAI Conference 2017 Conference Paper

Reactive Versus Anticipative Decision Making in a Novel Gift-Giving Game

• Elias Fern‡ndez Domingos
• Juan Burguillo
• Tom Lenaerts

Evolutionary game theory focuses on the fitness differences between simple discrete or probabilistic strategies to explain the evolution of particular decision-making behavior within strategic situations. Although this approach has provided substantial insights into the presence of fairness or generosity in gift-giving games, it does not fully resolve the question of which cognitive mechanisms are required to produce the choices observed in experiments. One such mechanism that humans have acquired, is the capacity to anticipate. Prior work showed that forward-looking behavior, using a recurrent neural network to model the cognitive mechanism, are essential to produce the actions of human participants in behavioral experiments. In this paper, we evaluate whether this conclusion extends also to gift-giving games, more concretely, to a game that combines the dictator game with a partner selection process. The recurrent neural network model used here for dictators, allows them to reason about a best response to past actions of the receivers (reactive model) or to decide which action will lead to a more successful outcome in the future (anticipatory model). We show for both models the decision dynamics while training, as well as the average behavior. We find that the anticipatory model is the only one capable of accounting for changes in the context of the game, a behavior also observed in experiments, expanding previous conclusions to this more sophisticated game.

AAMAS Conference 2017 Conference Paper

Social Manifestation of Guilt Leads to Stable Cooperation in Multi-Agent Systems

• Luis Moniz Pereira
• Tom Lenaerts
• Luis A. Martinez-Vaquero
• The Anh Han

Inspired by psychological and evolutionary studies, we present here theoretical models wherein agents have the potential to express guilt with the ambition to study the role of this emotion in the promotion of pro-social behaviour. To achieve this goal, analytical and numerical methods from evolutionary game theory are employed to identify the conditions for which enhanced cooperation emerges within the context of the iterated prisoners dilemma. Guilt is modelled explicitly as two features, i. e. a counter that keeps track of the number of transgressions and a threshold that dictates when alleviation (through for instance apology and self-punishment) is required for an emotional agent. Such an alleviation introduces an effect on the payoff of the agent experiencing guilt. We show that when the system consists of agents that resolve their guilt without considering the co-player’s attitude towards guilt alleviation then cooperation does not emerge. In that case those guilt prone agents are easily dominated by agents expressing no guilt or having no incentive to alleviate the guilt they experience. When, on the other hand, the guilt prone focal agent requires that guilt only needs to be alleviated when guilt alleviation is also manifested by a defecting co-player, then cooperation may thrive. This observation remains consistent for a generalised model as is discussed in this article. In summary, our analysis provides important insights into the design of multi-agent and cognitive agent systems where the inclusion of guilt modelling can improve agents’ cooperative behaviour and overall benefit.

JAAMAS Journal 2016 Journal Article

Evolution of commitment and level of participation in public goods games

• The Anh Han
• Luís Moniz Pereira
• Tom Lenaerts

Abstract Before engaging in a group venture agents may require commitments from other members in the group, and based on the level of acceptance (participation) they can then decide whether it is worthwhile joining the group effort. Here, we show in the context of public goods games and using stochastic evolutionary game theory modelling, which implies imitation and mutation dynamics, that arranging prior commitments while imposing a minimal participation when interacting in groups induces agents to behave cooperatively. Our analytical and numerical results show that if the cost of arranging the commitment is sufficiently small compared to the cost of cooperation, commitment arranging behavior is frequent, leading to a high level of cooperation in the population. Moreover, an optimal participation level emerges depending both on the dilemma at stake and on the cost of arranging the commitment. Namely, the harsher the common good dilemma is, and the costlier it becomes to arrange the commitment, the more participants should explicitly commit to the agreement to ensure the success of the joint venture. Furthermore, considering that commitment deals may last for more than one encounter, we show that commitment proposers can be lenient in case of short-term agreements, yet should be strict in case of long-term interactions.

IJCAI Conference 2013 Conference Paper

Evolution of Common-Pool Resources and Social Welfare in Structured Populations

• Jean-Sébastien Lerat
• The Anh Han
• Tom Lenaerts

The Common-pool resource (CPR) game is a social dilemma where agents have to decide how to consume a shared CPR. Either they each take their cut, completely destroying the CPR, or they restrain themselves, gaining less immediate profit but sustaining the resource and future profit. When no consumption takes place the CPR simply grows to its carrying capacity. As such, this dilemma provides a framework to study the evolution of social consumption strategies and the sustainability of resources, whose size adjusts dynamically through consumption and their own implicit population dynamics. The present study provides for the first time a detailed analysis of the evolutionary dynamics of consumption strategies in finite populations, focusing on the interplay between the resource levels and preferred consumption strategies. We show analytically which restrained consumers survive in relation to the growth rate of the resources and how this affects the resources’ carrying capacity. Second, we show that population structures affect the sustainability of the resources and social welfare in the population. Current results provide an initial insight into the complexity of the CPR game, showing potential for a variety of different studies in the context of social welfare and resource sustainability.

IJCAI Conference 2013 Conference Paper

Why Is It So Hard to Say Sorry? Evolution of Apology with Commitments in the Iterated Prisoner's Dilemma

• The Anh Han
• Luís Moniz Pereira
• Francisco C. Santos
• Tom Lenaerts

When making a mistake, individuals can apologize to secure further cooperation, even if the apology is costly. Similarly, individuals arrange commitments to guarantee that an action such as a cooperative one is in the others’ best interest, and thus will be carried out to avoid eventual penalties for commitment failure. Hence, both apology and commitment should go side by side in behavioral evolution. Here we provide a computational model showing that apologizing acts are rare in non-committed interactions, especially whenever cooperation is very costly, and that arranging prior commitments can considerably increase the frequency of such behavior. In addition, we show that in both cases, with or without commitments, apology works only if it is sincere, i. e. costly enough. Most interestingly, our model predicts that individuals tend to use much costlier apology in committed relationships than otherwise, because it helps better identify free-riders such as fake committers: ‘commitments bring about sincerity’. Furthermore, we show that this strategy of apology supported by commitments outperforms the famous existent strategies of the iterated Prisoner’s Dilemma.

AAMAS Conference 2003 Conference Paper

A Selection-Mutation model for Q-learning in Multi-Agent Systems

• Karl Tuyls
• Katja Verbeeck
• Tom Lenaerts