A.J. Sobey

Mohammad Divband Soorati, Enrico H. Gerding, Enrico Marchioni et al.

The Agents, Interaction and Complexity research group at the University of Southampton has a long track record of research in multiagent systems (MAS). We have made substantial scientific contributions across learning in MAS, game-theoretic techniques for coordinating agent systems, and formal methods for representation and reasoning. We highlight key results achieved by the group and elaborate on recent work and open research challenges in developing trustworthy autonomous systems and deploying human-centred AI systems that aim to support societal good.

Sizhe Yuen, Thomas H. G. Ezard, Adam J. Sobey

Hundreds of Evolutionary Computation approaches have been reported. From an evolutionary perspective they focus on two fundamental mechanisms: cultural inheritance in Swarm Intelligence and genetic inheritance in Evolutionary Algorithms. Contemporary evolutionary biology looks beyond genetic inheritance, proposing a so-called ``Extended Evolutionary Synthesis''. Many concepts from the Extended Evolutionary Synthesis have been left out of Evolutionary Computation as interest has moved toward specific implementations of the same general mechanisms. One such concept is epigenetic inheritance, which is increasingly considered central to evolutionary thinking. Epigenetic mechanisms allow quick non- or partially-genetic adaptations to environmental changes. Dynamic multi-objective optimisation problems represent similar circumstances to the natural world where fitness can be determined by multiple objectives (traits), and the environment is constantly changing. This paper asks if the advantages that epigenetic inheritance provide in the natural world are replicated in dynamic multi-objective optimisation problems. Specifically, an epigenetic blocking mechanism is applied to a state-of-the-art multi-objective genetic algorithm, MOEA/D-DE, and its performance is compared on three sets of dynamic test functions, FDA, JY, and UDF. The mechanism shows improved performance on 12 of the 16 test problems, providing initial evidence that more algorithms should explore the wealth of epigenetic mechanisms seen in the natural world.

Olaf Lipinski, Adam J. Sobey, Federico Cerutti et al.

Emergent communication studies the development of language between autonomous agents, aiming to improve understanding of natural language evolution and increase communication efficiency. While temporal aspects of language have been considered in computational linguistics, there has been no research on temporal references in emergent communication. This paper addresses this gap, by exploring how agents communicate about temporal relationships. We analyse three potential influences for the emergence of temporal references: environmental, external, and architectural changes. Our experiments demonstrate that altering the loss function is insufficient for temporal references to emerge; rather, architectural changes are necessary. However, a minimal change in agent architecture, using a different batching method, allows the emergence of temporal references. This modified design is compared with the standard architecture in a temporal referential games environment, which emphasises temporal relationships. The analysis indicates that over 95\% of the agents with the modified batching method develop temporal references, without changes to their loss function. We consider temporal referencing necessary for future improvements to the agents' communication efficiency, yielding a closer to optimal coding as compared to purely compositional languages. Our readily transferable architectural insights provide the basis for their incorporation into other emergent communication settings.

John Birkbeck, Adam Sobey, Federico Cerutti et al.

Reinforcement learning (RL) agents are costly to train and fragile to environmental changes. They often perform poorly when there are many changing tasks, prohibiting their widespread deployment in the real world. Many Lifelong RL agent designs have been proposed to mitigate issues such as catastrophic forgetting or demonstrate positive characteristics like forward transfer when change occurs. However, no prior work has established whether the impact on agent performance can be predicted from the change itself. Understanding this relationship will help agents proactively mitigate a change's impact for improved learning performance. We propose Change-Induced Regret Proxy (CHIRP) metrics to link change to agent performance drops and use two environments to demonstrate a CHIRP's utility in lifelong learning. A simple CHIRP-based agent achieved $48\%$ higher performance than the next best method in one benchmark and attained the best success rates in 8 of 10 tasks in a second benchmark which proved difficult for existing lifelong RL agents.

Cody Kommers, Drew Hemment, Maria Antoniak et al.

This position paper argues that large language models (LLMs) can make cultural context, and therefore human meaning, legible at an unprecedented scale in AI-based sociotechnical systems. We argue that such systems have previously been unable to represent human meaning because they rely on thin descriptions: numerical representations that enforce standardization and therefore strip human activity of the cultural context that gives it meaning. By contrast, scholars in the humanities and qualitative social sciences have developed frameworks for representing meaning through thick description: verbal representations that accommodate heterogeneity and retain contextual information needed to represent human meaning. While these methods can effectively codify meaning, they are difficult to deploy at scale. However, the verbal capabilities of LLMs now provide a means of (at least partially) automating the generation and processing of thick descriptions, potentially overcoming this bottleneck. We argue that the problem of rendering human meaning legible is not just about selecting better metrics, but about developing new representational formats (based on thick description). We frame this as a crucial direction for the application of generative AI and identify five key challenges: preserving context, maintaining interpretive pluralism, integrating perspectives based on lived experience and critical distance, distinguishing qualitative content from quantitative magnitude, and acknowledging meaning as dynamic rather than static. Furthermore, we suggest that thick description has the potential to serve as a unifying framework to address a number of emerging concerns about the difficulties of representing culture in (or using) LLMs.

Wan Wang, Haiyan Wang, Adam J. Sobey

Problem definition: Supply chains are constantly evolving networks. Reinforcement learning is increasingly proposed as a solution to provide optimal control of these networks. Academic/practical: However, learning in continuously varying environments remains a challenge in the reinforcement learning literature.Methodology: This paper therefore seeks to address whether agents can control varying supply chain problems, transferring learning between environments that require different strategies and avoiding catastrophic forgetting of tasks that have not been seen in a while. To evaluate this approach, two state-of-the-art Reinforcement Learning (RL) algorithms are compared: an actor-critic learner, Proximal Policy Optimisation(PPO), and a Recurrent Proximal Policy Optimisation (RPPO), PPO with a Long Short-Term Memory(LSTM) layer, which is showing popularity in online learning environments. Results: First these methods are compared on six sets of environments with varying degrees of stochasticity. The results show that more lean strategies adopted in Batch environments are different from those adopted in Stochastic environments with varying products. The methods are also compared on various continuous supply chain scenarios, where the PPO agents are shown to be able to adapt through continuous learning when the tasks are similar but show more volatile performance when changing between the extreme tasks. However, the RPPO, with an ability to remember histories, is able to overcome this to some extent and takes on a more realistic strategy. Managerial implications: Our results provide a new perspective on the continuously varying supply chain, the cooperation and coordination of agents are crucial for improving the overall performance in uncertain and semi-continuous non-stationary supply chain environments without the need to retrain the environment as the demand changes.

Kryspin Varys, Federico Cerutti, Adam Sobey et al.

Our society is governed by a set of norms which together bring about the values we cherish such as safety, fairness or trustworthiness. The goal of value-alignment is to create agents that not only do their tasks but through their behaviours also promote these values. Many of the norms are written as laws or rules (legal / safety norms) but even more remain unwritten (social norms). Furthermore, the techniques used to represent these norms also differ. Safety / legal norms are often represented explicitly, for example, in some logical language while social norms are typically learned and remain hidden in the parameter space of a neural network. There is a lack of approaches in the literature that could combine these various norm representations into a single algorithm. We propose a novel method that integrates these norms into the reinforcement learning process. Our method monitors the agent's compliance with the given norms and summarizes it in a quantity we call the agent's reputation. This quantity is used to weigh the received rewards to motivate the agent to become value-aligned. We carry out a series of experiments including a continuous state space traffic problem to demonstrate the importance of the written and unwritten norms and show how our method can find the value-aligned policies. Furthermore, we carry out ablations to demonstrate why it is better to combine these two groups of norms rather than using either separately.

Sizhe Yuen, Ting Su, Ziyang Wang et al. · cmu

A question-answering (QA) system is to search suitable answers within a knowledge base. Current QA systems struggle with queries requiring complex reasoning or real-time knowledge integration. They are often supplemented with retrieval techniques on a data source such as Retrieval-Augmented Generation (RAG). However, RAG continues to face challenges in handling complex reasoning and logical connections between multiple sources of information. A novel approach for enhancing Large Language Models (LLMs) in knowledge-intensive QA tasks is presented through the automated generation of context-based QA pairs. This methodology leverages LLMs to create fine-tuning data, reducing reliance on human labelling and improving model comprehension and reasoning capabilities. The proposed system includes an automated QA generator and a model fine-tuner, evaluated using perplexity, ROUGE, BLEU, and BERTScore. Comprehensive experiments demonstrate improvements in logical coherence and factual accuracy, with implications for developing adaptable Artificial Intelligence (AI) systems. Mistral-7b-v0.3 outperforms Llama-3-8b with BERT F1, BLEU, and ROUGE scores 0.858, 0.172, and 0.260 of for the LLM generated QA pairs compared to scores of 0.836, 0.083, and 0.139 for the human annotated QA pairs.

Olaf Lipinski, Adam J. Sobey, Federico Cerutti et al.

Effective communication requires the ability to refer to specific parts of an observation in relation to others. While emergent communication literature shows success in developing various language properties, no research has shown the emergence of such positional references. This paper demonstrates how agents can communicate about spatial relationships within their observations. The results indicate that agents can develop a language capable of expressing the relationships between parts of their observation, achieving over 90% accuracy when trained in a referential game which requires such communication. Using a collocation measure, we demonstrate how the agents create such references. This analysis suggests that agents use a mixture of non-compositional and compositional messages to convey spatial relationships. We also show that the emergent language is interpretable by humans. The translation accuracy is tested by communicating with the receiver agent, where the receiver achieves over 78% accuracy using parts of this lexicon, confirming that the interpretation of the emergent language was successful.

Sizhe Yuen, Thomas H. G. Ezard, Adam J. Sobey

Evolutionary Computation is a group of biologically inspired algorithms used to solve complex optimisation problems. It can be split into Evolutionary Algorithms, which take inspiration from genetic inheritance, and Swarm Intelligence algorithms, that take inspiration from cultural inheritance. However, recent developments have focused on computational or mathematical adaptions, leaving their biological roots behind. This has left much of the modern evolutionary literature relatively unexplored. To understand which evolutionary mechanisms have been considered, and which have been overlooked, this paper breaks down successful bio-inspired algorithms under a contemporary biological framework based on the Extended Evolutionary Synthesis, an extension of the classical, genetics focussed, Modern Synthesis. The analysis shows that Darwinism and the Modern Synthesis have been incorporated into Evolutionary Computation but that the Extended Evolutionary Synthesis has been broadly ignored beyond:cultural inheritance, incorporated in the sub-set of Swarm Intelligence algorithms, evolvability, through CMA-ES, and multilevel selection, through Multi-Level Selection Genetic Algorithm. The framework shows a missing gap in epigenetic inheritance for Evolutionary Computation, despite being a key building block in modern interpretations of how evolution occurs. Epigenetic inheritance can explain fast adaptation, without changes in an individual's genotype, by allowing biological organisms to self-adapt quickly to environmental cues, which, increases the speed of convergence while maintaining stability in changing environments. This leaves a diverse range of biologically inspired mechanisms as low hanging fruit that should be explored further within Evolutionary Computation.

David M. Bossens, Adam J. Sobey

A long-standing challenge in artificial intelligence is lifelong reinforcement learning, where learners are given many tasks in sequence and must transfer knowledge between tasks while avoiding catastrophic forgetting. Policy reuse and other multi-policy reinforcement learning techniques can learn multiple tasks but may generate many policies. This paper presents two novel contributions, namely 1) Lifetime Policy Reuse, a model-agnostic policy reuse algorithm that avoids generating many policies by optimising a fixed number of near-optimal policies through a combination of policy optimisation and adaptive policy selection; and 2) the task capacity, a measure for the maximal number of tasks that a policy can accurately solve. Comparing two state-of-the-art base-learners, the results demonstrate the importance of Lifetime Policy Reuse and task capacity based pre-selection on an 18-task partially observable Pacman domain and a Cartpole domain of up to 125 tasks.

A. I. Parkes, A. J. Sobey, D. A. Hudson

Artificial intelligence is applied in a range of sectors, and is relied upon for decisions requiring a high level of trust. For regression methods, trust is increased if they approximate the true input-output relationships and perform accurately outside the bounds of the training data. But often performance off-test-set is poor, especially when data is sparse. This is because the conditional average, which in many scenarios is a good approximation of the `ground truth', is only modelled with conventional Minkowski-r error measures when the data set adheres to restrictive assumptions, with many real data sets violating these. To combat this there are several methods that use prior knowledge to approximate the `ground truth'. However, prior knowledge is not always available, and this paper investigates how error measures affect the ability for a regression method to model the `ground truth' in these scenarios. Current error measures are shown to create an unhelpful bias and a new error measure is derived which does not exhibit this behaviour. This is tested on 36 representative data sets with different characteristics, showing that it is more consistent in determining the `ground truth' and in giving improved predictions in regions beyond the range of the training data.

P. A. Grudniewski, A. J. Sobey

In practical optimisation the dominant characteristics of the problem are often not known prior. Therefore, there is a need to develop general solvers as it is not always possible to tailor a specialised approach to each application. The hybrid form of Multi-Level Selection Genetic Algorithm (MLSGA) already shows good performance on range of problems due to its diversity-first approach, which is rare among Evolutionary Algorithms. To increase the generality of its performance this paper proposes a distinct set of co-evolutionary mechanisms, which defines co-evolution as competition between collectives rather than individuals. This distinctive approach to co-evolutionary provides less regular communication between sub-populations and different fitness definitions between individuals and collectives. This encourages the collectives to act more independently creating a unique sub-regional search, leading to the development of co-evolutionary MLSGA (cMLSGA). To test this methodology nine genetic algorithms are selected to generate several variants of cMLSGA, which incorporates these approaches at the individual level. The new mechanisms are tested on over 100 different functions and benchmarked against the 9 state-of-the-art competitors in order to find the best general solver. The results show that the diversity of co-evolutionary approaches is more important than their individual performances. This allows the selection of two competing algorithms that improve the generality of cMLSGA, without large loss of performance on any specific problem type. When compared to the state-of-the-art, the proposed methodology is the most universal and robust, leading to an algorithm more likely to solve complex problems with limited knowledge about the search space.