A neural model of hierarchical reinforcement learning

Autonomous robot navigation based on a hierarchical cognitive model

We propose a hierarchical cognitive navigation model (HCNM) to improve the self-learning and self-adaptive ability of mobile robots in unknown and complex environments. The HCNM model adopts the divide and conquers approach by dividing the path planning task into different levels of sub-tasks in complex environments and solves each sub-task in a smaller state subspace to decrease the state space dimensions. The HCNM model imitates animal asymptotic properties through the study of thermodynamic processes and designs a cognitive learning algorithm to achieve online optimum search strategies. We prove that the learning algorithm designed ensures that the cognitive model can converge to the optimal behavior path with probability one. Robot navigation is studied on the basis of the cognitive process. The experimental results show that the HCNM model has strong adaptability in unknown and environment, and the navigation path is clearer and the convergence time is better. Among them, the convergence time of HCNM model is 25 s, which is 86.5% lower than that of HRLM model. The HCNM model studied in this paper adopts a hierarchical structure, which reduces the learning difficulty and accelerates the learning speed in the unknown environment.

Hierarchical Reinforcement Learning, Sequential Behavior, and the Dorsal Frontostriatal System

To effectively behave within ever-changing environments, biological agents must learn and act at varying hierarchical levels such that a complex task may be broken down into more tractable subtasks. Hierarchical reinforcement learning (HRL) is a computational framework that provides an understanding of this process by combining sequential actions into one temporally extended unit called an option. However, there are still open questions within the HRL framework, including how options are formed and how HRL mechanisms might be realized within the brain. In this review, we propose that the existing human motor sequence literature can aid in understanding both of these questions. We give specific emphasis to visuomotor sequence learning tasks such as the discrete sequence production task and the M × N (M steps × N sets) task to understand how hierarchical learning and behavior manifest across sequential action tasks as well as how the dorsal cortical-subcortical circuitry could support this kind of behavior. This review highlights how motor chunks within a motor sequence can function as HRL options. Furthermore, we aim to merge findings from motor sequence literature with reinforcement learning perspectives to inform experimental design in each respective subfield.

Spiking Neural Networks and Their Applications: A Review

The past decade has witnessed the great success of deep neural networks in various domains. However, deep neural networks are very resource-intensive in terms of energy consumption, data requirements, and high computational costs. With the recent increasing need for the autonomy of machines in the real world, e.g., self-driving vehicles, drones, and collaborative robots, exploitation of deep neural networks in those applications has been actively investigated. In those applications, energy and computational efficiencies are especially important because of the need for real-time responses and the limited energy supply. A promising solution to these previously infeasible applications has recently been given by biologically plausible spiking neural networks. Spiking neural networks aim to bridge the gap between neuroscience and machine learning, using biologically realistic models of neurons to carry out the computation. Due to their functional similarity to the biological neural network, spiking neural networks can embrace the sparsity found in biology and are highly compatible with temporal code. Our contributions in this work are: (i) we give a comprehensive review of theories of biological neurons; (ii) we present various existing spike-based neuron models, which have been studied in neuroscience; (iii) we detail synapse models; (iv) we provide a review of artificial neural networks; (v) we provide detailed guidance on how to train spike-based neuron models; (vi) we revise available spike-based neuron frameworks that have been developed to support implementing spiking neural networks; (vii) finally, we cover existing spiking neural network applications in computer vision and robotics domains. The paper concludes with discussions of future perspectives.

Eye movements reveal spatiotemporal dynamics of visually-informed planning in navigation

Goal-oriented navigation is widely understood to depend upon internal maps. Although this may be the case in many settings, humans tend to rely on vision in complex, unfamiliar environments. To study the nature of gaze during visually-guided navigation, we tasked humans to navigate to transiently visible goals in virtual mazes of varying levels of difficulty, observing that they took near-optimal trajectories in all arenas. By analyzing participants' eye movements, we gained insights into how they performed visually-informed planning. The spatial distribution of gaze revealed that environmental complexity mediated a striking trade-off in the extent to which attention was directed towards two complimentary aspects of the world model: the reward location and task-relevant transitions. The temporal evolution of gaze revealed rapid, sequential prospection of the future path, evocative of neural replay. These findings suggest that the spatiotemporal characteristics of gaze during navigation are significantly shaped by the unique cognitive computations underlying real-world, sequential decision making.

Neuromorphic Engineering Needs Closed-Loop Benchmarks

Neuromorphic engineering aims to build (autonomous) systems by mimicking biological systems. It is motivated by the observation that biological organisms—from algae to primates—excel in sensing their environment, reacting promptly to their perils and opportunities. Furthermore, they do so more resiliently than our most advanced machines, at a fraction of the power consumption. It follows that the performance of neuromorphic systems should be evaluated in terms of real-time operation, power consumption, and resiliency to real-world perturbations and noise using task-relevant evaluation metrics. Yet, following in the footsteps of conventional machine learning, most neuromorphic benchmarks rely on recorded datasets that foster sensing accuracy as the primary measure for performance. Sensing accuracy is but an arbitrary proxy for the actual system's goal—taking a good decision in a timely manner. Moreover, static datasets hinder our ability to study and compare closed-loop sensing and control strategies that are central to survival for biological organisms. This article makes the case for a renewed focus on closed-loop benchmarks involving real-world tasks. Such benchmarks will be crucial in developing and progressing neuromorphic Intelligence. The shift towards dynamic real-world benchmarking tasks should usher in richer, more resilient, and robust artificially intelligent systems in the future.

Fine-Grained Video Captioning via Graph-based Multi-Granularity Interaction Learning

Learning to generate continuous linguistic descriptions for multi-subject interactive videos in great details has particular applications in team sports auto-narrative. In contrast to traditional video caption, this task is more challenging as it requires simultaneous modeling of fine-grained individual actions, uncovering of spatio-temporal dependency structures of frequent group interactions, and then accurate mapping of these complex interaction details into long and detailed commentary. To explicitly address these challenges, we propose a novel framework Graph-based Learning for Multi-Granularity Interaction Representation (GLMGIR) for fine-grained team sports auto-narrative task. A multi-granular interaction modeling module is proposed to extract among-subjects' interactive actions in a progressive way for encoding both intra- and inter-team interactions. Based on the above multi-granular representations, a multi-granular attention module is developed to consider action/event descriptions of multiple spatio-temporal resolutions. Both modules are integrated seamlessly and work in a collaborative way to generate the final narrative. In the meantime, to facilitate reproducible research, we collect a new video dataset from YouTube.com called Sports Video Narrative dataset (SVN). It is a novel direction as it contains 6K team sports videos (i.e., NBA basketball games) with 10K ground-truth narratives(e.g., sentences). Furthermore, as previous metrics such as METEOR (i.e., used in coarse-grained video caption task) DO NOT cope with fine-grained sports narrative task well, we hence develop a novel evaluation metric named Fine-grained Captioning Evaluation (FCE), which measures how accurate the generated linguistic description reflects fine-grained action details as well as the overall spatio-temporal interactional structure. Extensive experiments on our SVN dataset have demonstrated the effectiveness of the proposed framework for fine-grained team sports video auto-narrative.

Multimodality carotid plaque tissue characterization and classification in the artificial intelligence paradigm: a narrative review for stroke application

Cardiovascular disease (CVD) is one of the leading causes of morbidity and mortality in the United States of America and globally. Carotid arterial plaque, a cause and also a marker of such CVD, can be detected by various non-invasive imaging modalities such as magnetic resonance imaging (MRI), computer tomography (CT), and ultrasound (US). Characterization and classification of carotid plaque-type in these imaging modalities, especially into symptomatic and asymptomatic plaque, helps in the planning of carotid endarterectomy or stenting. It can be challenging to characterize plaque components due to (I) partial volume effect in magnetic resonance imaging (MRI) or (II) varying Hausdorff values in plaque regions in CT, and (III) attenuation of echoes reflected by the plaque during US causing acoustic shadowing. Artificial intelligence (AI) methods have become an indispensable part of healthcare and their applications to the non-invasive imaging technologies such as MRI, CT, and the US. In this narrative review, three main types of AI models (machine learning, deep learning, and transfer learning) are analyzed when applied to MRI, CT, and the US. A link between carotid plaque characteristics and the risk of coronary artery disease is presented. With regard to characterization, we review tools and techniques that use AI models to distinguish carotid plaque types based on signal processing and feature strengths. We conclude that AI-based solutions offer an accurate and robust path for tissue characterization and classification for carotid artery plaque imaging in all three imaging modalities. Due to cost, user-friendliness, and clinical effectiveness, AI in the US has dominated the most.

Discovery of hierarchical representations for efficient planning

We propose that humans spontaneously organize environments into clusters of states that support hierarchical planning, enabling them to tackle challenging problems by breaking them down into sub-problems at various levels of abstraction. People constantly rely on such hierarchical presentations to accomplish tasks big and small-from planning one's day, to organizing a wedding, to getting a PhD-often succeeding on the very first attempt. We formalize a Bayesian model of hierarchy discovery that explains how humans discover such useful abstractions. Building on principles developed in structure learning and robotics, the model predicts that hierarchy discovery should be sensitive to the topological structure, reward distribution, and distribution of tasks in the environment. In five simulations, we show that the model accounts for previously reported effects of environment structure on planning behavior, such as detection of bottleneck states and transitions. We then test the novel predictions of the model in eight behavioral experiments, demonstrating how the distribution of tasks and rewards can influence planning behavior via the discovered hierarchy, sometimes facilitating and sometimes hindering performance. We find evidence that the hierarchy discovery process unfolds incrementally across trials. Finally, we propose how hierarchy discovery and hierarchical planning might be implemented in the brain. Together, these findings present an important advance in our understanding of how the brain might use Bayesian inference to discover and exploit the hidden hierarchical structure of the environment.

Autonomous Functional Movements in a Tendon-Driven Limb via Limited Experience

Robots will become ubiquitously useful only when they can use few attempts to teach themselves to perform different tasks, even with complex bodies and in dynamical environments. Vertebrates, in fact, use sparse trial-and-error to learn multiple tasks despite their intricate tendon-driven anatomies-which are particularly hard to control because they are simultaneously nonlinear, under-determined, and over-determined. We demonstrate-for the first time in simulation and hardware-how a model-free, open-loop approach allows few-shot autonomous learning to produce effective movements in a 3-tendon 2-joint limb. We use a short period of motor babbling (to create an initial inverse map) followed by building functional habits by reinforcing high-reward behavior and refinements of the inverse map in a movement's neighborhood. This biologically-plausible algorithm, which we call G2P (General-to-Particular), can potentially enable quick, robust and versatile adaptation in robots as well as shed light on the foundations of the enviable functional versatility of organisms.

Feature-specific prediction errors and surprise across macaque fronto-striatal circuits

To adjust expectations efficiently, prediction errors need to be associated with the precise features that gave rise to the unexpected outcome, but this credit assignment may be problematic if stimuli differ on multiple dimensions and it is ambiguous which feature dimension caused the outcome. Here, we report a potential solution: neurons in four recorded areas of the anterior fronto-striatal networks encode prediction errors that are specific to feature values of different dimensions of attended multidimensional stimuli. The most ubiquitous prediction error occurred for the reward-relevant dimension. Feature-specific prediction error signals a) emerge on average shortly after non-specific prediction error signals, b) arise earliest in the anterior cingulate cortex and later in dorsolateral prefrontal cortex, caudate and ventral striatum, and c) contribute to feature-based stimulus selection after learning. Thus, a widely-distributed feature-specific eligibility trace may be used to update synaptic weights for improved feature-based attention.

In order to adjust expectations efficiently, prediction errors need to be associated with the features that gave rise to the unexpected outcome. Here, the authors show that neurons in anterior fronto-striatal networks encode prediction errors that are specific to feature values of different stimulus dimensions.

How Organisms Gained Causal Independence and How It Might Be Quantified

Two broad features are jointly necessary for autonomous agency: organisational closure and the embodiment of an objective-function providing a ‘goal’: so far only organisms demonstrate both. Organisational closure has been studied (mostly in abstract), especially as cell autopoiesis and the cybernetic principles of autonomy, but the role of an internalised ‘goal’ and how it is instantiated by cell signalling and the functioning of nervous systems has received less attention. Here I add some biological ‘flesh’ to the cybernetic theory and trace the evolutionary development of step-changes in autonomy: (1) homeostasis of organisationally closed systems; (2) perception-action systems; (3) action selection systems; (4) cognitive systems; (5) memory supporting a self-model able to anticipate and evaluate actions and consequences. Each stage is characterised by the number of nested goal-directed control-loops embodied by the organism, summarised as will-nestedness N. Organism tegument, receptor/transducer system, mechanisms of cellular and whole-organism re-programming and organisational integration, all contribute to causal independence.

CONCLUSION

organisms are cybernetic phenomena whose identity is created by the information structure of the highest level of causal closure (maximum N), which has increased through evolution, leading to increased causal independence, which might be quantifiable by ‘Integrated Information Theory’ measures.