arXiv subjects

Kenji Doya

Publications and source records attributed to Kenji Doya.

At least 19 recordsLinked to original sources

arXiv2026

The Dually Flat Geometry of Planning as Inference

We present an alternative characterization of the occupancy measure of reinforcement learning, obtained by embedding the planning criterion into the dynamics through a resetting planning process. Its stationary measure, which we term visitation measure, is the object on which the information geometry of decision making is most naturally expressed. The achievable visitation measures form a dually flat statistical manifold whose two affine charts are the visitation probabilities and the log-policies, dual under the conditional entropy. This structure makes planning-as-inference generalize from linear rewards to nonlinear functionals of the visitation, each iterate solved by one natural-gradient step, and gives the temporal-difference error the interpretation of a marginal-utility estimate. We develop the geometry and its consequences for reinforcement learning and theoretical neuroscience.

cs.AI↗

arXiv2026

Extended predictive coding framework as variational free-energy minimisation under exponential-family assumption

The sensory cortices of the brain perform perceptual inference efficiently through their complex networks of neurons. One of the theoretical accounts of this process is the free-energy principle (FEP), which postulates that the brain performs variational Bayesian inference. Pioneering studies have shown that FEP can correspond to the predictive coding (PC) hypothesis under the Gaussian assumption and Laplace approximation. However, PC-based implementations of FEP within such a limited Gaussian regime have failed to capture several properties of biological neural networks, such as nonlinearity and heterogeneity of input--output properties within a network, and the biological implausibility of negative firing rates. This study shows that, when a broader class of probability distributions, namely the exponential family of distributions (EFD), is assumed for the variational posterior and prior, these missing characteristics are exhibited within the network, maintaining the FEP--PC correspondence up to the second cumulant of the posterior. We also show that the proposed model can be trained by biologically plausible local plasticity rules. Our results enrich the explanatory power of FEP regarding neural dynamics involved in perception as variational inference.

q-bio.NC↗

arXiv2025

Lag Operator SSMs: A Geometric Framework for Structured State Space Modeling

Structured State Space Models (SSMs), which are at the heart of the recently popular Mamba architecture, are powerful tools for sequence modeling. However, their theoretical foundation relies on a complex, multistage process of continuous-time modeling and subsequent discretization, which can obscure intuition. We introduce a direct, first-principles framework for constructing discrete-time SSMs that is both flexible and modular. Our approach is based on a novel lag operator, which geometrically derives the discrete-time recurrence by measuring how the system's basis functions undergo what we call a domain expansion from one timestep to the next. The resulting state matrices are computed via a single inner product involving this operator, enabling a modular design space for creating novel SSMs by flexibly combining different basis functions and time-warping schemes. To validate our framework, we demonstrate that a specific instance exactly recovers the recurrence of the influential HiPPO model. Numerical simulations confirm our derivation, providing new theoretical tools for designing flexible and robust sequence models.

cs.LG↗

arXiv2025

Curiosity-Driven Development of Action and Language in Robots Through Self-Exploration

Infants acquire language with generalization from minimal experience, whereas large language models require billions of training tokens. What underlies efficient development in humans? We investigated this problem through experiments wherein robotic agents learn to perform actions associated with imperative sentences (e.g., push red cube) via curiosity-driven self-exploration. Our approach amortizes active inference using Q-learning, enabling intrinsically motivated developmental learning. The simulations reveal key findings corresponding to observations in developmental psychology. i) Generalization improves drastically as the scale of compositional elements increases. ii) Curiosity-driven exploration enables faster learning. iii) Rote pairing of sentences and actions precedes compositional generalization. iv) Exception-handling induces U-shaped developmental performance, a pattern like representational redescription in child language learning. These results suggest that curiosity-driven active inference accounts for how intrinsically motivated sensorimotor-linguistic learning supports scalable compositional generalization and exception handling in humans and artificial agents.

stat.ML↗

arXiv2025

Evolution of Fear and Social Rewards in Prey-Predator Relationship

Fear is a critical brain function that enables us to learn to avoid danger via reinforcement learning (RL). While many researchers have argued that fear has evolved to escape predators, how varying predatory pressures have shaped fear and other rewards, including positive social rewards for collective grouping, remains an open question. In this study, we investigate the relationship between predatory pressure and fear using an evolutionary simulation of RL agents with evolving rewards. In our simulation, prey and predator RL agents co-evolve their reward functions, including visual rewards for observing prey and predators. While fear-like negative visual rewards for predators often evolved in prey, we also observed cases in which positive rewards for both predators and prey evolved, the latter serving as a social reward for collective grouping. A comparison between different environmental conditions revealed that stronger predator hunting capability promoted stronger fear reward, while less food supply promoted more negative social reward. Moreover, fear did not evolve in response to static pitfalls with non-lethal damage, suggesting that actively hunting predators played an important role in its evolution. These results highlight the special role of predators in the diverse evolution of fear and social rewards.

q-bio.PE↗

arXiv2024

Estimating Exoplanet Mass using Machine Learning on Incomplete Datasets

The exoplanet archive is an incredible resource of information on the properties of discovered extrasolar planets, but statistical analysis has been limited by the number of missing values. One of the most informative bulk properties is planet mass, which is particularly challenging to measure with more than 70\% of discovered planets with no measured value. We compare the capabilities of five different machine learning algorithms that can utilize multidimensional incomplete datasets to estimate missing properties for imputing planet mass. The results are compared when using a partial subset of the archive with a complete set of six planet properties, and where all planet discoveries are leveraged in an incomplete set of six and eight planet properties. We find that imputation results improve with more data even when the additional data is incomplete, and allows a mass prediction for any planet regardless of which properties are known. Our favored algorithm is the newly developed $k$NN$\times$KDE, which can return a probability distribution for the imputed properties. The shape of this distribution can indicate the algorithm's level of confidence, and also inform on the underlying demographics of the exoplanet population. We demonstrate how the distributions can be interpreted with a series of examples for planets where the discovery was made with either the transit method, or radial velocity method. Finally, we test the generative capability of the $k$NN$\times$KDE to create a large synthetic population of planets based on the archive, and identify potential categories of planets from groups of properties in the multidimensional space. All codes are Open Source.

astro-ph.EP↗

arXiv2024

Evolution of Rewards for Food and Motor Action by Simulating Birth and Death

The reward system is one of the fundamental drivers of animal behaviors and is critical for survival and reproduction. Despite its importance, the problem of how the reward system has evolved is underexplored. In this paper, we try to replicate the evolution of biologically plausible reward functions and investigate how environmental conditions affect evolved rewards' shape. For this purpose, we developed a population-based decentralized evolutionary simulation framework, where agents maintain their energy level to live longer and produce more children. Each agent inherits its reward function from its parent subject to mutation and learns to get rewards via reinforcement learning throughout its lifetime. Our results show that biologically reasonable positive rewards for food acquisition and negative rewards for motor action can evolve from randomly initialized ones. However, we also find that the rewards for motor action diverge into two modes: largely positive and slightly negative. The emergence of positive motor action rewards is surprising because it can make agents too active and inefficient in foraging. In environments with poor and poisonous foods, the evolution of rewards for less important foods tends to be unstable, while rewards for normal foods are still stable. These results demonstrate the usefulness of our simulation environment and energy-dependent birth and death model for further studies of the origin of reward systems.

cs.NE↗

arXiv2024

Intrinsic Rewards for Exploration without Harm from Observational Noise: A Simulation Study Based on the Free Energy Principle

In Reinforcement Learning (RL), artificial agents are trained to maximize numerical rewards by performing tasks. Exploration is essential in RL because agents must discover information before exploiting it. Two rewards encouraging efficient exploration are the entropy of action policy and curiosity for information gain. Entropy is well-established in literature, promoting randomized action selection. Curiosity is defined in a broad variety of ways in literature, promoting discovery of novel experiences. One example, prediction error curiosity, rewards agents for discovering observations they cannot accurately predict. However, such agents may be distracted by unpredictable observational noises known as curiosity traps. Based on the Free Energy Principle (FEP), this paper proposes hidden state curiosity, which rewards agents by the KL divergence between the predictive prior and posterior probabilities of latent variables. We trained six types of agents to navigate mazes: baseline agents without rewards for entropy or curiosity, and agents rewarded for entropy and/or either prediction error curiosity or hidden state curiosity. We find entropy and curiosity result in efficient exploration, especially both employed together. Notably, agents with hidden state curiosity demonstrate resilience against curiosity traps, which hinder agents with prediction error curiosity. This suggests implementing the FEP may enhance the robustness and generalization of RL models, potentially aligning the learning processes of artificial and biological agents.

cs.LG↗

arXiv2023

Numerical Data Imputation for Multimodal Data Sets: A Probabilistic Nearest-Neighbor Kernel Density Approach

Numerical data imputation algorithms replace missing values by estimates to leverage incomplete data sets. Current imputation methods seek to minimize the error between the unobserved ground truth and the imputed values. But this strategy can create artifacts leading to poor imputation in the presence of multimodal or complex distributions. To tackle this problem, we introduce the $k$NN$\times$KDE algorithm: a data imputation method combining nearest neighbor estimation ($k$NN) and density estimation with Gaussian kernels (KDE). We compare our method with previous data imputation methods using artificial and real-world data with different data missing scenarios and various data missing rates, and show that our method can cope with complex original data structure, yields lower data imputation errors, and provides probabilistic estimates with higher likelihood than current methods. We release the code in open-source for the community: https://github.com/DeltaFloflo/knnxkde

stat.ML↗

arXiv2023

Habits and goals in synergy: a variational Bayesian framework for behavior

How to behave efficiently and flexibly is a central problem for understanding biological agents and creating intelligent embodied AI. It has been well known that behavior can be classified as two types: reward-maximizing habitual behavior, which is fast while inflexible; and goal-directed behavior, which is flexible while slow. Conventionally, habitual and goal-directed behaviors are considered handled by two distinct systems in the brain. Here, we propose to bridge the gap between the two behaviors, drawing on the principles of variational Bayesian theory. We incorporate both behaviors in one framework by introducing a Bayesian latent variable called "intention". The habitual behavior is generated by using prior distribution of intention, which is goal-less; and the goal-directed behavior is generated by the posterior distribution of intention, which is conditioned on the goal. Building on this idea, we present a novel Bayesian framework for modeling behaviors. Our proposed framework enables skill sharing between the two kinds of behaviors, and by leveraging the idea of predictive coding, it enables an agent to seamlessly generalize from habitual to goal-directed behavior without requiring additional training. The proposed framework suggests a fresh perspective for cognitive science and embodied AI, highlighting the potential for greater integration between habitual and goal-directed behaviors.

cs.LG↗

arXiv2021

Goal-Directed Planning by Reinforcement Learning and Active Inference

What is the difference between goal-directed and habitual behavior? We propose a novel computational framework of decision making with Bayesian inference, in which everything is integrated as an entire neural network model. The model learns to predict environmental state transitions by self-exploration and generating motor actions by sampling stochastic internal states ${z}$. Habitual behavior, which is obtained from the prior distribution of ${z}$, is acquired by reinforcement learning. Goal-directed behavior is determined from the posterior distribution of ${z}$ by planning, using active inference which optimizes the past, current and future ${z}$ by minimizing the variational free energy for the desired future observation constrained by the observed sensory sequence. We demonstrate the effectiveness of the proposed framework by experiments in a sensorimotor navigation task with camera observations and continuous motor actions.

cs.LG↗

arXiv2021

Canonical Cortical Circuits and the Duality of Bayesian Inference and Optimal Control

The duality of sensory inference and motor control has been known since the 1960s and has recently been recognized as the commonality in computations required for the posterior distributions in Bayesian inference and the value functions in optimal control. Meanwhile, an intriguing question about the brain is why the entire neocortex shares a canonical six-layer architecture while its posterior and anterior halves are engaged in sensory processing and motor control, respectively. Here we consider the hypothesis that the sensory and motor cortical circuits implement the dual computations for Bayesian inference and optimal control, or perceptual and value-based decision making, respectively. We first review the classic duality of inference and control in linear quadratic systems and then review the correspondence between dynamic Bayesian inference and optimal control. Based on the architecture of the canonical cortical circuit, we explore how different cortical neurons may represent variables and implement computations.

q-bio.NC↗

arXiv2021

A Whole Brain Probabilistic Generative Model: Toward Realizing Cognitive Architectures for Developmental Robots

Building a humanlike integrative artificial cognitive system, that is, an artificial general intelligence (AGI), is the holy grail of the artificial intelligence (AI) field. Furthermore, a computational model that enables an artificial system to achieve cognitive development will be an excellent reference for brain and cognitive science. This paper describes an approach to develop a cognitive architecture by integrating elemental cognitive modules to enable the training of the modules as a whole. This approach is based on two ideas: (1) brain-inspired AI, learning human brain architecture to build human-level intelligence, and (2) a probabilistic generative model(PGM)-based cognitive system to develop a cognitive system for developmental robots by integrating PGMs. The development framework is called a whole brain PGM (WB-PGM), which differs fundamentally from existing cognitive architectures in that it can learn continuously through a system based on sensory-motor information. In this study, we describe the rationale of WB-PGM, the current status of PGM-based elemental cognitive modules, their relationship with the human brain, the approach to the integration of the cognitive modules, and future challenges. Our findings can serve as a reference for brain studies. As PGMs describe explicit informational relationships between variables, this description provides interpretable guidance from computational sciences to brain science. By providing such information, researchers in neuroscience can provide feedback to researchers in AI and robotics on what the current models lack with reference to the brain. Further, it can facilitate collaboration among researchers in neuro-cognitive sciences as well as AI and robotics.

cs.AI↗

arXiv2020

Forward and inverse reinforcement learning sharing network weights and hyperparameters

This paper proposes model-free imitation learning named Entropy-Regularized Imitation Learning (ERIL) that minimizes the reverse Kullback-Leibler (KL) divergence. ERIL combines forward and inverse reinforcement learning (RL) under the framework of an entropy-regularized Markov decision process. An inverse RL step computes the log-ratio between two distributions by evaluating two binary discriminators. The first discriminator distinguishes the state generated by the forward RL step from the expert's state. The second discriminator, which is structured by the theory of entropy regularization, distinguishes the state-action-next-state tuples generated by the learner from the expert ones. One notable feature is that the second discriminator shares hyperparameters with the forward RL, which can be used to control the discriminator's ability. A forward RL step minimizes the reverse KL estimated by the inverse RL step. We show that minimizing the reverse KL divergence is equivalent to finding an optimal policy. Our experimental results on MuJoCo-simulated environments and vision-based reaching tasks with a robotic arm show that ERIL is more sample-efficient than the baseline methods. We apply the method to human behaviors that perform a pole-balancing task and describe how the estimated reward functions show how every subject achieves her goal.

cs.LG↗

arXiv2019

Variational Recurrent Models for Solving Partially Observable Control Tasks

In partially observable (PO) environments, deep reinforcement learning (RL) agents often suffer from unsatisfactory performance, since two problems need to be tackled together: how to extract information from the raw observations to solve the task, and how to improve the policy. In this study, we propose an RL algorithm for solving PO tasks. Our method comprises two parts: a variational recurrent model (VRM) for modeling the environment, and an RL controller that has access to both the environment and the VRM. The proposed algorithm was tested in two types of PO robotic control tasks, those in which either coordinates or velocities were not observable and those that require long-term memorization. Our experiments show that the proposed algorithm achieved better data efficiency and/or learned more optimal policy than other alternative approaches in tasks in which unobserved states cannot be inferred from raw observations in a simple manner.

cs.LG↗

arXiv2019

Optimization and Validation of Diffusion MRI-based Fiber Tracking with Neural Tracer Data as a Reference

Diffusion-weighted magnetic resonance imaging (dMRI) allows non-invasive investigation of whole-brain connectivity, which can potentially help to reveal the brain's global network architecture and abnormalities involved in neurological and mental disorders. However, the reliability of connection inferences from dMRI-based fiber tracking is still debated, due to low sensitivity, dominance of false positives, and inaccurate and incomplete reconstruction of long-range connections. Furthermore, parameters of tracking algorithms are typically tuned in a heuristic way, which leaves room for manipulation of an intended result. Here we propose a data-driven framework to optimize and validate parameters of dMRI-based fiber-tracking algorithms using neural tracer data as a reference. Japan's Brain/MINDS Project provides invaluable datasets containing both dMRI and neural tracer data from the same primates. We considered four criteria for goodness of fiber tracking: distance-weighted coverage, true/false positive ratio, projection coincidence, and commissural passage, applied using a multi-objective optimization algorithm. We implemented a variant of non-dominated sorting genetic algorithm II (NSGA-II) to optimize five major parameters of a global fiber-tracking algorithm over multiple brain samples in parallel. Using optimized parameters compared to the default parameters, dMRI-based fiber tracking performance was significantly improved, while minimizing false positives and impossible cross-hemisphere connections. Parameters optimized for 10 tracer injection sites showed good generalization capability for other brain samples. These results demonstrate the importance of data-driven adjustment of fiber-tracking algorithms and support the validity of dMRI-based tractography, if appropriate adjustments are employed.

physics.med-ph↗

arXiv2019

MarmoNet: a pipeline for automated projection mapping of the common marmoset brain from whole-brain serial two-photon tomography

Understanding the connectivity in the brain is an important prerequisite for understanding how the brain processes information. In the Brain/MINDS project, a connectivity study on marmoset brains uses two-photon microscopy fluorescence images of axonal projections to collect the neuron connectivity from defined brain regions at the mesoscopic scale. The processing of the images requires the detection and segmentation of the axonal tracer signal. The objective is to detect as much tracer signal as possible while not misclassifying other background structures as the signal. This can be challenging because of imaging noise, a cluttered image background, distortions or varying image contrast cause problems. We are developing MarmoNet, a pipeline that processes and analyzes tracer image data of the common marmoset brain. The pipeline incorporates state-of-the-art machine learning techniques based on artificial convolutional neural networks (CNN) and image registration techniques to extract and map all relevant information in a robust manner. The pipeline processes new images in a fully automated way. This report introduces the current state of the tracer signal analysis part of the pipeline.

eess.IV↗

arXiv2019

Gap-Increasing Policy Evaluation for Efficient and Noise-Tolerant Reinforcement Learning

In real-world applications of reinforcement learning (RL), noise from inherent stochasticity of environments is inevitable. However, current policy evaluation algorithms, which plays a key role in many RL algorithms, are either prone to noise or inefficient. To solve this issue, we introduce a novel policy evaluation algorithm, which we call Gap-increasing RetrAce Policy Evaluation (GRAPE). It leverages two recent ideas: (1) gap-increasing value update operators in advantage learning for noise-tolerance and (2) off-policy eligibility trace in Retrace algorithm for efficient learning. We provide detailed theoretical analysis of the new algorithm that shows its efficiency and noise-tolerance inherited from Retrace and advantage learning. Furthermore, our analysis shows that GRAPE's learning is significantly efficient than that of a simple learning-rate-based approach while keeping the same level of noise-tolerance. We applied GRAPE to control problems and obtained experimental results supporting our theoretical analysis.

cs.LG↗