Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices through online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often “unintended” systemic outcomes. This paper introduces human-AI coevolution as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., scientific, legal and socio-political.

Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices through online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often “unintended” systemic outcomes. This paper introduces human-AI coevolution as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., scientific, legal and socio-political.

A Self-Organising Temporal Pooling (SOTP) network has been shown to be capable of forming declarative parallel representations of sequential events and chunking these events without supervision. However, such a network currently cannot take these declarative representations and execute the associated sequence; it is strictly a one-way sequence chunker and encoder. We present a modified model that can convert the lateral weights of the trained network into a primacy gradient usable by methods such as Competitive Queueing (CQ), thus granting the network a method for executing learned sequences. The resulting model has several benefits over traditional CQ. We further present an advanced method of executing sequences via SOTP itself, resulting in less error than CQ whilst being more flexible in replaying sequences from datasets with variable sequence lengths.

Temporal Pooling (TP) is a recent technique for processing temporal events by forming declarative representations of the complete sequences. In this paper, we examine and extend the functionality of the existing TP algorithm from the Hierarchical Temporal Memory (HTM) framework and introduce the Self-Organising Temporal Pooling (SOTP) architecture. The SOTP draws together the Merge Self-Organising Map (MSOM) and a new algorithm termed the Modified Temporal Pooler (MTP) to produce functionality not yet seen in a TP. This new architecture is shown to be capable of producing stable declarative representations of sequences that activate consistently for each item in that sequence.

Due to the huge size of the market for interactive games, game controllers are very cheap. Because of this they are a popular choice as input devices for research in user interfaces. The Software Development Kit that comes with each controller also makes it appealing compared with the effort of creating an original input device. We present a natural user interface for 3D interaction that uses a game controller. Our design is based on five core characteristics: i) It uses cheap hardware as an interface; ii) It implements a 6DoF interaction; iii) It allows 3D manipulation of an object; iv) We test the interface by experiment; and v) We compare the natural interface with a standard mouse interface doing a similar task. Results from the experiment showed that our natural interface is easy to use and has a faster completion times compared with the standard modeling interface.

A significant feature of spiking neural networks with varying connection delays, such as those in the brain, is the existence of strongly connected groups of neurons known as polychronous neural groups (PNGs). Polychronous groups are found in large numbers in these networks and are proposed by Izhikevich (2006a) to provide a neural basis for representation and memory. When exposed to a familiar stimulus, spiking neural networks produce consistencies in the spiking output data that are the hallmarks of PNG activation. Previous methods for studying the PNG activation response to stimuli have been limited by the template-based methods used to identify PNG activation. In this letter, we outline a new method that overcomes these difficulties by establishing for the first time a probabilistic interpretation of PNG activation. We then demonstrate the use of this method by investigating the claim that PNGs might provide the foundation of a representational system.

This paper describes an animated conversational agent called Kare which integrates a talking head interface with a linguistically motivated human-machine dialogue system. The agent has a range of nonverbal behaviors, which involve a mixture of machine vision, computer animation and natural language processing techniques. The system's architecture couples the agent's nonverbal communicative processes very tightly to its model of verbal interaction. We discuss several consequences of this architecture, in particular the ability to use different nonverbal dialogue management signals when speaking different languages.

We argue in this article that many common adverbial phrases generally taken to signal a discourse relation between syntactically connected units within discourse structure instead work anaphorically to contribute relational meaning, with only indirect dependence on discourse structure. This allows a simpler discourse structure to provide scaffolding for compositional semantics and reveals multiple ways in which the relational meaning conveyed by adverbial connectives can interact with that associated with discourse structure. We conclude by sketching out a lexicalized grammar for discourse that facilitates discourse interpretation as a product of compositional rules, anaphor resolution, and inference.

Generating text in a hypermedia environment places different demands on a text generation system than occurs in non-interactive environments. This paper describes some of these demands, then shows how the architecture of one text generation system, ILEX, has been shaped by them. The architecture is described in terms of the levels of linguistic representation used, and the processes which map between them. Particular attention is paid to the processes of content selection and text structuring.

This paper outlines a text generation system suited to a large class of information sources, relational databases. We focus on one aspect of the problem: the additional information which needs to be specified to produce reasonable text quality when generating from relational databases. We outline how databases need to be prepared, and then describe various types of domain semantics which can be used to improve text quality.

We will demonstrate the ILEX system, a system which dynamically generates descriptions of database objects for the web, adapting the description to the discourse context and user type. Among other improvements in version 3, the system now generates from relational databases, and this demonstration will focus on this ability. We will also show how incremental extensions to the domain semantics improve the quality of the text produced.

In this paper, a new framework is proposed for defining the semantics of discourse relations and for defining the semantics of the utterances that relations link together. The proposal is to define relations in terms of the operation of an algorithm simulating the mental state of an agent interacting with the world. The algorithm interleaves perception, theorem proving, and action: The denotation of a complex utterance containing a relation between two simpler utterances is taken to be the description of the operation of the algorithm during the time interval identified by their Reichenbachian reference times. This proposal is presented in detail for two discourse relations. Its potential application in the treatment of mood, tense, aspect, and dialogue structure is also discussed in very general terms.

We show that discourse structure need not bear the full burden of conveying discourse relations by showing that many of them can be explained nonstructurally in terms of the grounding of anaphoric presuppositions (Van der Sandt, 1992). This simplifies discourse structure, while still allowing the realisation of a full range of discourse relations. This is achieved using the same semantic machinery used in deriving clause-level semantics.

We discuss a task requiring the coherent presentation of heterogeneous information about objects recorded in electronic catalogues. We consider the advantages of combining hypermedia delivery with natural language generation technology, so as to allow us to view a session with such a system as a coherent conversation or dialogue. We describe two prototype systems we have built which make use of these combined techniques, and focus on those aspects of the systems which attempt to provide coherence. Although the techniques themselves are not novel, their combination is relatively recent, and promises to help forge useful tools for accomplishing our specific information retrieval task.

Dynamic hypertext' is hypertext which is automatically generated at the point of need.A number of NLG systems have now been developed to operate within a hypertext environment; and now that these systems are becoming widely available on the World Wide Web, it is useful to take stock of how well-equipped NLG technology is to work in this new domain.A generation system in a hypertext environment faces a specific set of requirements; here, we discuss those requirements, and the resources that can be provided to help meet them.Examples are drawn from a number of systems, including our own prototype, ILEX-0.We conclude by indicating that the major benefit of such systems could be in the way that they combine flexibility with the illusion of user control.