Contemporary machine learning paradigm excels in statistical data analysis, solving problems that classical AI couldn't. However, it faces key limitations, such as a lack of integration with planning, incomprehensible internal structure, and inability to learn continually. We present the initial design for an AI system, Agential AI (AAI), in principle operating independently or on top of statistical methods, designed to overcome these issues. AAI's core is a learning method that models temporal dynamics with guarantees of completeness, minimality, and continual learning, using component-level variation and selection to learn the structure of the environment. It integrates this with a behavior algorithm that plans on a learned model and encapsulates high-level behavior patterns. Preliminary experiments on a simple environment show AAI's effectiveness and potential.

For the robots to achieve a desired behavior, we can program them directly, train them, or give them an innate driver that makes the robots themselves desire the targeted behavior. With the minimal surprise approach, we implant in our robots the desire to make their world predictable. Here, we apply minimal surprise to collective construction. Simulated robots push blocks in a 2D torus grid world. In two variants of our experiment we either allow for emergent behaviors or predefine the expected environment of the robots. In either way, we evolve robot behaviors that move blocks to structure their environment and make it more predictable. The resulting controllers can be applied in collective construction by robots.

Effective action abstraction is crucial in tackling challenges associated with large action spaces in Imperfect Information Extensive-Form Games (IIEFGs). However, due to the vast state space and computational complexity in IIEFGs, existing methods often rely on fixed abstractions, resulting in sub-optimal performance. In response, we introduce RL-CFR, a novel reinforcement learning (RL) approach for dynamic action abstraction. RL-CFR builds upon our innovative Markov Decision Process (MDP) formulation, with states corresponding to public information and actions represented as feature vectors indicating specific action abstractions. The reward is defined as the expected payoff difference between the selected and default action abstractions. RL-CFR constructs a game tree with RL-guided action abstractions and utilizes counterfactual regret minimization (CFR) for strategy derivation. Impressively, it can be trained from scratch, achieving higher expected payoff without increased CFR solving time. In experiments on Heads-up No-limit Texas Hold'em, RL-CFR outperforms ReBeL's replication and Slumbot, demonstrating significant win-rate margins of $64\pm 11$ and $84\pm 17$ mbb/hand, respectively.

Graph neural networks are popular architectures for graph machine learning, based on iterative computation of node representations of an input graph through a series of invariant transformations. A large class of graph neural networks follow a standard message-passing paradigm: at every layer, each node state is updated based on an aggregate of messages from its neighborhood. In this work, we propose a novel framework for training graph neural networks, where every node is viewed as a player that can choose to either 'listen', 'broadcast', 'listen and broadcast', or to 'isolate'. The standard message propagation scheme can then be viewed as a special case of this framework where every node 'listens and broadcasts' to all neighbors. Our approach offers a more flexible and dynamic message-passing paradigm, where each node can determine its own strategy based on their state, effectively exploring the graph topology while learning. We provide a theoretical analysis of the new message-passing scheme which is further supported by an extensive empirical analysis on a synthetic dataset and on real-world datasets.

Predictive coding has emerged as a prominent model of how the brain learns through predictions, anticipating the importance accorded to predictive learning in recent AI architectures such as transformers. Here we propose a new framework for predictive coding called active predictive coding which can learn hierarchical world models and solve two radically different open problems in AI: (1) how do we learn compositional representations, e.g., part-whole hierarchies, for equivariant vision? and (2) how do we solve large-scale planning problems, which are hard for traditional reinforcement learning, by composing complex action sequences from primitive policies? Our approach exploits hypernetworks, self-supervised learning and reinforcement learning to learn hierarchical world models that combine task-invariant state transition networks and task-dependent policy networks at multiple abstraction levels. We demonstrate the viability of our approach on a variety of vision datasets (MNIST, FashionMNIST, Omniglot) as well as on a scalable hierarchical planning problem. Our results represent, to our knowledge, the first demonstration of a unified solution to the part-whole learning problem posed by Hinton, the nested reference frames problem posed by Hawkins, and the integrated state-action hierarchy learning problem in reinforcement learning.

We introduce Active Predictive Coding Networks (APCNs), a new class of neural networks that solve a major problem posed by Hinton and others in the fields of artificial intelligence and brain modeling: how can neural networks learn intrinsic reference frames for objects and parse visual scenes into part-whole hierarchies by dynamically allocating nodes in a parse tree? APCNs address this problem by using a novel combination of ideas: (1) hypernetworks are used for dynamically generating recurrent neural networks that predict parts and their locations within intrinsic reference frames conditioned on higher object-level embedding vectors, and (2) reinforcement learning is used in conjunction with backpropagation for end-to-end learning of model parameters. The APCN architecture lends itself naturally to multi-level hierarchical learning and is closely related to predictive coding models of cortical function. Using the MNIST, Fashion-MNIST and Omniglot datasets, we demonstrate that APCNs can (a) learn to parse images into part-whole hierarchies, (b) learn compositional representations, and (c) transfer their knowledge to unseen classes of objects. With their ability to dynamically generate parse trees with part locations for objects, APCNs offer a new framework for explainable AI that leverages advances in deep learning while retaining interpretability and compositionality.

We address a state-of-the-art reinforcement learning (RL) control approach to automatically configure robotic prosthesis impedance parameters to enable end-to-end, continuous locomotion intended for transfemoral amputee subjects. Specifically, our actor-critic based RL provides tracking control of a robotic knee prosthesis to mimic the intact knee profile. This is a significant advance from our previous RL based automatic tuning of prosthesis control parameters which have centered on regulation control with a designer prescribed robotic knee profile as the target. In addition to presenting the complete tracking control algorithm based on direct heuristic dynamic programming (dHDP), we provide an analytical framework for the tracking controller with constrained inputs. We show that our proposed tracking control possesses several important properties, such as weight convergence of the learning networks, Bellman (sub)optimality of the cost-to-go value function and control input, and practical stability of the human-robot system under input constraint. We further provide a systematic simulation of the proposed tracking control using a realistic human-robot system simulator, the OpenSim, to emulate how the dHDP enables level ground walking, walking on different terrains and at different paces. These results show that our proposed dHDP based tracking control is not only theoretically suitable, but also practically useful.

Many studies have been made to improve the AVs' ability of environment recognition and vehicle control, while the attention paid to decision making is not enough though the decision algorithms so far are very preliminary. Therefore, a framework of the decision-making training and learning is put forward in this paper. It consists of two parts: the deep reinforcement learning (DRL) training program and the high-fidelity virtual simulation environment. Then the basic microscopic behavior, car-following (CF), is trained within this framework. In addition, theoretical analysis and experiments were conducted on setting reward function for accelerating training using DRL. The results show that on the premise of driving comfort, the efficiency of the trained AV increases 7.9% compared to the classical traffic model, intelligent driver model (IDM). Later on, on a more complex three-lane section, we trained the integrated model combines both CF and lane-changing (LC) behavior, the average speed further grows 2.4%. It indicates that our framework is effective for AV's decision-making learning. Keywords: Automated vehicle; Decision making; Deep reinforcement learning; Reward function 1. Introduction The automated vehicles have captured the public attention in recent years, especially after Google announced its automated driving program in 2010, for its advantages of alleviating the traffic congestion, liberating drivers' attention and conserving energy. The tasks involved in achieving autonomous driving can be divided into three modules: environment recognition, decision making and vehicle control. Among them, the vehicle control has no obvious differences between AV and manual driven vehicle.

Advances in deep reinforcement learning have allowed autonomous agents to perform well on Atari games, often outperforming humans, using only raw pixels to make their decisions. However, most of these games take place in 2D environments that are fully observable to the agent. In this paper, we present the first architecture to tackle 3D environments in first-person shooter games, that involve partially observable states. Typically, deep reinforcement learning methods only utilize visual input for training. We present a method to augment these models to exploit game feature information such as the presence of enemies or items, during the training phase. Our model is trained to simultaneously learn these features along with minimizing a Q-learning objective, which is shown to dramatically improve the training speed and performance of our agent. Our architecture is also modularized to allow different models to be independently trained for different phases of the game. We show that the proposed architecture substantially outperforms built-in AI agents of the game as well as average humans in deathmatch scenarios.

Advances in deep reinforcement learning have allowed autonomous agents to perform well on Atari games, often outperforming humans, using only raw pixels to make their decisions. However, most of these games take place in 2D environments that are fully observable to the agent. In this paper, we present the first architecture to tackle 3D environments in first-person shooter games, that involve partially observable states. Typically, deep reinforcement learning methods only utilize visual input for training. We present a method to augment these models to exploit game feature information such as the presence of enemies or items, during the training phase. Our model is trained to simultaneously learn these features along with minimizing a Q-learning objective, which is shown to dramatically improve the training speed and performance of our agent. Our architecture is also modularized to allow different models to be independently trained for different phases of the game. We show that the proposed architecture substantially outperforms built-in AI agents of the game as well as average humans in deathmatch scenarios.