arXiv subjects

Takashi Ikegami

Publications and source records attributed to Takashi Ikegami.

At least 37 records · Page 2Linked to original sources

arXiv2023

From Text to Motion: Grounding GPT-4 in a Humanoid Robot "Alter3"

We report the development of Alter3, a humanoid robot capable of generating spontaneous motion using a Large Language Model (LLM), specifically GPT-4. This achievement was realized by integrating GPT-4 into our proprietary android, Alter3, thereby effectively grounding the LLM with Alter's bodily movement. Typically, low-level robot control is hardware-dependent and falls outside the scope of LLM corpora, presenting challenges for direct LLM-based robot control. However, in the case of humanoid robots like Alter3, direct control is feasible by mapping the linguistic expressions of human actions onto the robot's body through program code. Remarkably, this approach enables Alter3 to adopt various poses, such as a 'selfie' stance or 'pretending to be a ghost,' and generate sequences of actions over time without explicit programming for each body part. This demonstrates the robot's zero-shot learning capabilities. Additionally, verbal feedback can adjust poses, obviating the need for fine-tuning. A video of Alter3's generated motions is available at https://tnoinkwms.github.io/ALTER-LLM/

cs.RO↗

arXiv2023

Implementation of Lenia as a Reaction-Diffusion System

The relationship between reaction-diffusion (RD) systems, characterized by continuous spatiotemporal states, and cellular automata (CA), marked by discrete spatiotemporal states, remains poorly understood. This paper delves into this relationship through an examination of a recently developed CA known as Lenia. We demonstrate that asymptotic Lenia, a variant of Lenia, can be comprehensively described by differential equations, and, unlike the original Lenia, it is independent of time-step ticks. Further, we establish that this formulation is mathematically equivalent to a generalization of the kernel-based Turing model (KT model). Stemming from these insights, we establish that asymptotic Lenia can be replicated by an RD system composed solely of diffusion and spatially local reaction terms, resulting in the simulated asymptotic Lenia based on an RD system, or "RD Lenia". However, our RD Lenia cannot be construed as a chemical system since the reaction term fails to satisfy mass-action kinetics.

nlin.CG↗

arXiv2022

Organization of a Latent Space structure in VAE/GAN trained by navigation data

We present a novel artificial cognitive mapping system using generative deep neural networks, called variational autoencoder/generative adversarial network (VAE/GAN), which can map input images to latent vectors and generate temporal sequences internally. The results show that the distance of the predicted image is reflected in the distance of the corresponding latent vector after training. This indicates that the latent space is self-organized to reflect the proximity structure of the dataset and may provide a mechanism through which many aspects of cognition are spatially represented. The present study allows the network to internally generate temporal sequences that are analogous to the hippocampal replay/pre-play ability, where VAE produces only near-accurate replays of past experiences, but by introducing GANs, the generated sequences are coupled with instability and novelty.

cs.LG↗

arXiv2020

Spatial and Temporal Taylor's Law in 1-Dim Chaotic Maps

By using low-dimensional chaos maps, the power law relationship established between the sample mean and variance called Taylor's Law (TL) is studied. In particular, we aim to clarify the relationship between TL from the spatial ensemble (STL) and the temporal ensemble (TTL). Since the spatial ensemble corresponds to independent sampling from a stationary distribution, we confirm that STL is explained by the skewness of the distribution. The difference between TTL and STL is shown to be originated in the temporal correlation of a dynamics. In case of logistic and tent maps, the quadratic relationship in the mean and variance, called Bartlett's law, is found analytically. On the other hand, TTL in the Hassell model can be well explained by the chunk structure of the trajectory, whereas the TTL of the Ricker model have a different mechanism originated from the specific form of the map.

nlin.CD↗

arXiv2020

Neural Autopoiesis: Organizing Self-Boundary by Stimulus Avoidance in Biological and Artificial Neural Networks

Living organisms must actively maintain themselves in order to continue existing. Autopoiesis is a key concept in the study of living organisms, where the boundaries of the organism is not static by dynamically regulated by the system itself. To study the autonomous regulation of self-boundary, we focus on neural homeodynamic responses to environmental changes using both biological and artificial neural networks. Previous studies showed that embodied cultured neural networks and spiking neural networks with spike-timing dependent plasticity (STDP) learn an action as they avoid stimulation from outside. In this paper, as a result of our experiments using embodied cultured neurons, we find that there is also a second property allowing the network to avoid stimulation: if the agent cannot learn an action to avoid the external stimuli, it tends to decrease the stimulus-evoked spikes, as if to ignore the uncontrollable-input. We also show such a behavior is reproduced by spiking neural networks with asymmetric STDP. We consider that these properties are regarded as autonomous regulation of self and non-self for the network, in which a controllable-neuron is regarded as self, and an uncontrollable-neuron is regarded as non-self. Finally, we introduce neural autopoiesis by proposing the principle of stimulus avoidance.

cs.NE↗

arXiv2019

Predictive Coding as Stimulus Avoidance in Spiking Neural Networks

Predictive coding can be regarded as a function which reduces the error between an input signal and a top-down prediction. If reducing the error is equivalent to reducing the influence of stimuli from the environment, predictive coding can be regarded as stimulation avoidance by prediction. Our previous studies showed that action and selection for stimulation avoidance emerge in spiking neural networks through spike-timing dependent plasticity (STDP). In this study, we demonstrate that spiking neural networks with random structure spontaneously learn to predict temporal sequences of stimuli based solely on STDP.

cs.NE↗

arXiv2019

An Overview of Open-Ended Evolution: Editorial Introduction to the Open-Ended Evolution II Special Issue

Nature's spectacular inventiveness, reflected in the enormous diversity of form and function displayed by the biosphere, is a feature of life that distinguishes living most strongly from nonliving. It is, therefore, not surprising that this aspect of life should become a central focus of artificial life. We have known since Darwin that the diversity is produced dynamically, through the process of evolution; this has led life's creative productivity to be called Open-Ended Evolution (OEE) in the field. This article introduces the second of two special issues on current research in OEE and provides an overview of the contents of both special issues. Most of the work was presented at a workshop on open-ended evolution that was held as a part of the 2018 Conference on Artificial Life in Tokyo, and much of it had antecedents in two previous workshops on open-ended evolution at artificial life conferences in Cancun and York. We present a simplified categorization of OEE and summarize progress in the field as represented by the articles in this special issue.

cs.NE↗

arXiv2019

Open-ended Evolution and a Mechanism of Novelties in Web Services

Analogous to living ecosystems in nature, web services form an artificial ecosystem consisting of many tags and their associated media, such as photographs, movies, and web pages created by human users. Concerning biological ecosystems, we regard tag as a species and human as a hidden environmental resource. We subsequently analyze the evolution of the web services, in particular social tagging systems, with respect to the self-organization of new tags. The evolution of new combinations of tags is analyzed as the open-ended evolution (OEE) index. The tag meaning is computed by the types of associated tags; tags that vary their meanings temporally exist. We argue that such tags are the examples of OEE.

cs.SI↗

arXiv2019

How to Make Swarms Open-Ended? Evolving Collective Intelligence Through a Constricted Exploration of Adjacent Possibles

We propose an approach of open-ended evolution via the simulation of swarm dynamics. In nature, swarms possess remarkable properties, which allow many organisms, from swarming bacteria to ants and flocking birds, to form higher-order structures that enhance their behavior as a group. Swarm simulations highlight three important factors to create novelty and diversity: (a) communication generates combinatorial cooperative dynamics, (b) concurrency allows for separation of timescales, and (c) complexity and size increases push the system towards transitions in innovation. We illustrate these three components in a model computing the continuous evolution of a swarm of agents. The results, divided in three distinct applications, show how emergent structures are capable of filtering information through the bottleneck of their memory, to produce meaningful novelty and diversity within their simulated environment.

cs.MA↗

arXiv2019

Learning by Stimulation Avoidance: A Principle to Control Spiking Neural Networks Dynamics

Learning based on networks of real neurons, and by extension biologically inspired models of neural networks, has yet to find general learning rules leading to widespread applications. In this paper, we argue for the existence of a principle allowing to steer the dynamics of a biologically inspired neural network. Using carefully timed external stimulation, the network can be driven towards a desired dynamical state. We term this principle "Learning by Stimulation Avoidance" (LSA). We demonstrate through simulation that the minimal sufficient conditions leading to LSA in artificial networks are also sufficient to reproduce learning results similar to those obtained in biological neurons by Shahaf and Marom [1]. We examine the mechanism's basic dynamics in a reduced network, and demonstrate how it scales up to a network of 100 neurons. We show that LSA has a higher explanatory power than existing hypotheses about the response of biological neural networks to external simulation, and can be used as a learning rule for an embodied application: learning of wall avoidance by a simulated robot. The surge in popularity of artificial neural networks is mostly directed to disembodied models of neurons with biologically irrelevant dynamics: to the authors' knowledge, this is the first work demonstrating sensory-motor learning with random spiking networks through pure Hebbian learning.

cs.NE↗

arXiv2017

Anomalous popularity growth in social tagging ecosystems

In social tagging systems, the diversity of tag vocabulary and the popularity of such tags continue to increase as they are exposed to selection pressure derived from our cognitive nature and cultural preferences. This is analogous to living ecosystems, where mutation and selection play a dominant role. Such population dynamism, which yields a scaling law, is mathematically modeled by a simple stochastic process---the Yule--Simon process, which describes how new words are introduced to the system and then grow. However, in actual web services, we have observed that a large fluctuation emerges in the popularity growth of individual tags that cannot be explained by the ordinary selection mechanism. We introduce a scaling factor to quantify the degree of the deviation in the popularity growth from the mean-field solution of the Yule--Simon process, and we discuss possible triggers of such anomalous popularity behavior.

physics.soc-ph↗

arXiv2017

How the nature of web services drives vocabulary creation in social tagging

A social tagging system allows users to add arbitrary strings, called "tags", on a shared resource to organize and manage information. The Yule--Simon process, which has shown the ability to capture the population dynamics of social tagging behavior, does not handle the mechanism of new vocabulary creation because it assumes that new vocabulary creation is a Poisson-like random process. In this research, we focus on the mechanism of vocabulary creation from the microscopic perspective and discuss whether it also follows the random process assumed in the Yule--Simon process. To capture the microscopic mechanism of vocabulary creation, we focus on the relationship between the number of tags used in the same entry and the local vocabulary creation rate. We find that the relationship is not the result of a simple random process, and differs between services. Furthermore, these differences depend on whether the user's tagging attitudes are private or open. These results provide the potential for a new index to identify the service's intrinsic nature.

cs.SI↗

arXiv2016

Towards information based spatiotemporal patterns as a foundation for agent representation in dynamical systems

We present some arguments why existing methods for representing agents fall short in applications crucial to artificial life. Using a thought experiment involving a fictitious dynamical systems model of the biosphere we argue that the metabolism, motility, and the concept of counterfactual variation should be compatible with any agent representation in dynamical systems. We then propose an information-theoretic notion of \emph{integrated spatiotemporal patterns} which we believe can serve as the basic building block of an agent definition. We argue that these patterns are capable of solving the problems mentioned before. We also test this in some preliminary experiments.

cs.AI↗

arXiv2014

Short-Term Memory Through Persistent Activity: Evolution of Self-Stopping and Self-Sustaining Activity in Spiking Neural Networks

Memories in the brain are separated in two categories: short-term and long-term memories. Long-term memories remain for a lifetime, while short-term ones exist from a few milliseconds to a few minutes. Within short-term memory studies, there is debate about what neural structure could implement it. Indeed, mechanisms responsible for long-term memories appear inadequate for the task. Instead, it has been proposed that short-term memories could be sustained by the persistent activity of a group of neurons. In this work, we explore what topology could sustain short-term memories, not by designing a model from specific hypotheses, but through Darwinian evolution in order to obtain new insights into its implementation. We evolved 10 networks capable of retaining information for a fixed duration between 2 and 11s. Our main finding has been that the evolution naturally created two functional modules in the network: one which sustains the information containing primarily excitatory neurons, while the other, which is responsible for forgetting, was composed mainly of inhibitory neurons. This demonstrates how the balance between inhibition and excitation plays an important role in cognition.

cs.NE↗

arXiv2014

Dynamic Homeostasis in Packet Switching Networks

In this study, we investigate the adaptation and robustness of a packet switching network (PSN), the fundamental architecture of the Internet. We claim that the adaptation introduced by a transmission control protocol (TCP) congestion control mechanism is interpretable as the self-organization of multiple attractors and stability to switch from one attractor to another. To discuss this argument quantitatively, we study the adaptation of the Internet by simulating a PSN using ns-2. Our hypothesis is that the robustness and fragility of the Internet can be attributed to the inherent dynamics of the PSN feedback mechanism called the congestion window size, or \textit{cwnd}. By varying the data input into the PSN system, we investigate the possible self-organization of attractors in cwnd temporal dynamics and discuss the adaptability and robustness of PSNs. The present study provides an example of Ashby's Law of Requisite Variety in action.

cs.NI↗

arXiv2014

Neural coordination can be enhanced by occasional interruption of normal firing patterns: A self-optimizing spiking neural network model

The state space of a conventional Hopfield network typically exhibits many different attractors of which only a small subset satisfy constraints between neurons in a globally optimal fashion. It has recently been demonstrated that combining Hebbian learning with occasional alterations of normal neural states avoids this problem by means of self-organized enlargement of the best basins of attraction. However, so far it is not clear to what extent this process of self-optimization is also operative in real brains. Here we demonstrate that it can be transferred to more biologically plausible neural networks by implementing a self-optimizing spiking neural network model. In addition, by using this spiking neural network to emulate a Hopfield network with Hebbian learning, we attempt to make a connection between rate-based and temporal coding based neural systems. Although further work is required to make this model more realistic, it already suggests that the efficacy of the self-optimizing process is independent from the simplifying assumptions of a conventional Hopfield network. We also discuss natural and cultural processes that could be responsible for occasional alteration of neural firing patterns in actual brains

nlin.AO↗

arXiv2014

Self-organization on social media: endo-exo bursts and baseline fluctuations

A salient dynamic property of social media is bursting behavior. In this paper, we study bursting behavior in terms of the temporal relation between a preceding baseline fluctuation and the successive burst response using a frequency time series of 3,000 keywords on Twitter. We found that there is a fluctuation threshold up to which the burst size increases as the fluctuation increases and that above the threshold, there appears a variety of burst sizes. We call this threshold the critical threshold. Investigating this threshold in relation to endogenous bursts and exogenous bursts based on peak ratio and burst size reveals that the bursts below this threshold are endogenously caused and above this threshold, exogenous bursts emerge. Analysis of the 3,000 keywords shows that all the nouns have both endogenous and exogenous origins of bursts and that each keyword has a critical threshold in the baseline fluctuation value to distinguish between the two. Having a threshold for an input value for activating the system implies that Twitter is an excitable medium. These findings are useful for characterizing how excitable a keyword is on Twitter and could be used, for example, to predict the response to particular information on social media.

cs.SI↗

arXiv2014

Embodied social interaction constitutes social cognition in pairs of humans: A minimalist virtual reality experiment

Scientists have traditionally limited the mechanisms of social cognition to one brain, but recent approaches claim that interaction also realizes cognitive work. Experiments under constrained virtual settings revealed that interaction dynamics implicitly guide social cognition. Here we show that embodied social interaction can be constitutive of agency detection and of experiencing another`s presence. Pairs of participants moved their "avatars" along an invisible virtual line and could make haptic contact with three identical objects, two of which embodied the other`s motions, but only one, the other`s avatar, also embodied the other`s contact sensor and thereby enabled responsive interaction. Co-regulated interactions were significantly correlated with identifications of the other`s avatar and reports of the clearest awareness of the other`s presence. These results challenge folk psychological notions about the boundaries of mind, but make sense from evolutionary and developmental perspectives: an extendible mind can offload cognitive work into its environment.

nlin.AO↗