The Option Keyboard (OK) was recently proposed as a method for transferring behavioral knowledge across tasks. OK transfers knowledge by adaptively combining subsets of known behaviors using Successor Features (SFs) and Generalized Policy Improvement (GPI). However, it relies on hand-designed state-features and task encodings which are cumbersome to design for every new environment. In this work, we propose the "Successor Features Keyboard" (SFK), which enables transfer with discovered state-features and task encodings. To enable discovery, we propose the "Categorical Successor Feature Approximator" (CSFA), a novel learning algorithm for estimating SFs while jointly discovering state-features and task encodings. With SFK and CSFA, we achieve the first demonstration of transfer with SFs in a challenging 3D environment where all the necessary representations are discovered. We first compare CSFA against other methods for approximating SFs and show that only CSFA discovers representations compatible with SF&GPI at this scale. We then compare SFK against transfer learning baselines and show that it transfers most quickly to long-horizon tasks.

While deep reinforcement learning (RL) has been demonstrated effective in solving complex control tasks, sample efficiency remains a key challenge due to the large amounts of data required for remarkable performance. Existing research explores the application of representation learning for data-efficient RL, e.g., learning predictive representations by predicting long-term future states. However, many existing methods do not fully exploit the structural information inherent in sequential state signals, which can potentially improve the quality of long-term decision-making but is difficult to discern in the time domain. To tackle this problem, we propose State Sequences Prediction via Fourier Transform (SPF), a novel method that exploits the frequency domain of state sequences to extract the underlying patterns in time series data for learning expressive representations efficiently. Specifically, we theoretically analyze the existence of structural information in state sequences, which is closely related to policy performance and signal regularity, and then propose to predict the Fourier transform of infinite-step future state sequences to extract such information. One of the appealing features of SPF is that it is simple to implement while not requiring storage of infinite-step future states as prediction targets. Experiments demonstrate that the proposed method outperforms several state-of-the-art algorithms in terms of both sample efficiency and performance.

Imitation Learning (IL) aims to discover a policy by minimizing the discrepancy between the agent's behavior and expert demonstrations. However, IL is susceptible to limitations imposed by noisy demonstrations from non-expert behaviors, presenting a significant challenge due to the lack of supplementary information to assess their expertise. In this paper, we introduce Self-Motivated Imitation LEarning (SMILE), a method capable of progressively filtering out demonstrations collected by policies deemed inferior to the current policy, eliminating the need for additional information. We utilize the forward and reverse processes of Diffusion Models to emulate the shift in demonstration expertise from low to high and vice versa, thereby extracting the noise information that diffuses expertise. Then, the noise information is leveraged to predict the diffusion steps between the current policy and demonstrators, which we theoretically demonstrate its equivalence to their expertise gap. We further explain in detail how the predicted diffusion steps are applied to filter out noisy demonstrations in a self-motivated manner and provide its theoretical grounds. Through empirical evaluations on MuJoCo tasks, we demonstrate that our method is proficient in learning the expert policy amidst noisy demonstrations, and effectively filters out demonstrations with expertise inferior to the current policy.

Agile and adaptive maneuvers such as fall recovery, high-speed turning, and sprinting in the wild are challenging for legged systems. We propose a Curricular Hindsight Reinforcement Learning (CHRL) that learns an end-to-end tracking controller that achieves powerful agility and adaptation for the legged robot. The two key components are (I) a novel automatic curriculum strategy on task difficulty and (ii) a Hindsight Experience Replay strategy adapted to legged locomotion tasks. We demonstrated successful agile and adaptive locomotion on a real quadruped robot that performed fall recovery autonomously, coherent trotting, sustained outdoor speeds up to 3.45 m/s, and tuning speeds up to 3.2 rad/s. This system produces adaptive behaviours responding to changing situations and unexpected disturbances on natural terrains like grass and dirt.

Financial markets analysis has been and remains a topic of intense research interest since the seminal work of Markowitz [1] detailing his theory on portfolio choice, for which he was awarded the Nobel Prize in 1990. The rapid advancements of Machine Learning (ML) and, more specifically those made in the field of Deep Learning (DL) and Deep Reinforcement Learning (DRL), further fueled interest in the field. Financial markets analysts began using ML-based techniques and combining them with their own knowledge of the field [2]. As early as 1992, Neural Networks (NNs) were already being used for equity index futures trading [3]. More recently, DL research in financial market analysis has focused on high frequency trading, i.e., an algorithmic financial trading method where high speeds and large volumes are the main characteristics. The kind of data used in works that focus on this type of trading include Limit Order Book (LOB) data [4] as well as candle data for assets such as FOREX or Cryptocurrencies [5]. Candle data contain the Open, High, Low and Close prices for assets in a requested frequency, e.g., at the minute or hour level. Price forecasting is a first step towards solving the very complex task of portfolio management, and has proved to be a sufficiently difficult problem to tackle itself. One way to sufficiently solve it is by transforming the problem into one of classification, i.e., predicting the price movement instead of its actual value in the next step [4].

Designing and deriving effective model-based reinforcement learning (MBRL) algorithms with a performance improvement guarantee is challenging, mainly attributed to the high coupling between model learning and policy optimization. Many prior methods that rely on return discrepancy to guide model learning ignore the impacts of model shift, which can lead to performance deterioration due to excessive model updates. Other methods use performance difference bound to explicitly consider model shift. However, these methods rely on a fixed threshold to constrain model shift, resulting in a heavy dependence on the threshold and a lack of adaptability during the training process. In this paper, we theoretically derive an optimization objective that can unify model shift and model bias and then formulate a fine-tuning process. This process adaptively adjusts the model updates to get a performance improvement guarantee while avoiding model overfitting. Based on these, we develop a straightforward algorithm USB-PO (Unified model Shift and model Bias Policy Optimization). Empirical results show that USB-PO achieves state-of-the-art performance on several challenging benchmark tasks.

This paper proposes a new cognitive model, acting as the main component of an AGI agent. The model is introduced in its mature intelligence state, and as an extension of previous models, DENN, and especially AKREM, by including operational models (frames/classes) and will. This model's core assumption is that cognition is about operating on accumulated knowledge, with the guidance of an appropriate will. Also, we assume that the actions, part of knowledge, are learning to be aligned with will, during the evolution phase that precedes the mature intelligence state. In addition, this model is mainly based on the duality principle in every known intelligent aspect, such as exhibiting both top-down and bottom-up model learning, generalization verse specialization, and more. Furthermore, a holistic approach is advocated for AGI designing, and cognition under constraints or efficiency is proposed, in the form of reusability and simplicity. Finally, reaching this mature state is described via a cognitive evolution from infancy to adulthood, utilizing a consolidation principle. The final product of this cognitive model is a dynamic operational memory of models and instances. Lastly, some examples and preliminary ideas for the evolution phase to reach the mature state are presented.

Recent years have witnessed significant advancements in offline reinforcement learning (RL), resulting in the development of numerous algorithms with varying degrees of complexity. While these algorithms have led to noteworthy improvements, many incorporate seemingly minor design choices that impact their effectiveness beyond core algorithmic advances. However, the effect of these design choices on established baselines remains understudied. In this work, we aim to bridge this gap by conducting a retrospective analysis of recent works in offline RL and propose ReBRAC, a minimalistic algorithm that integrates such design elements built on top of the TD3+BC method. We evaluate ReBRAC on 51 datasets with both proprioceptive and visual state spaces using D4RL and V-D4RL benchmarks, demonstrating its state-of-the-art performance among ensemble-free methods in both offline and offline-to-online settings. To further illustrate the efficacy of these design choices, we perform a large-scale ablation study and hyperparameter sensitivity analysis on the scale of thousands of experiments.

In multi-agent informative path planning (MAIPP), agents must collectively construct a global belief map of an underlying distribution of interest (e.g., gas concentration, light intensity, or pollution levels) over a given domain, based on measurements taken along their trajectory. They must frequently replan their path to balance the exploration of new areas with the exploitation of known high-interest areas, to maximize information gain within a predefined budget. Traditional approaches rely on reactive path planning conditioned on other agents' predicted future actions. However, as the belief is continuously updated, the predicted actions may not match the executed actions, introducing noise and reducing performance. We propose a decentralized, deep reinforcement learning (DRL) approach using an attention-based neural network, where agents optimize long-term individual and cooperative objectives by sharing their intent, represented as a distribution of medium-/long-term future positions obtained from their own policy. Intent sharing enables agents to learn to claim or avoid broader areas, while the use of attention mechanisms allows them to identify useful portions of imperfect predictions, maximizing cooperation even based on imperfect information. Our experiments compare the performance of our approach, its variants, and high-quality baselines across various MAIPP scenarios. We finally demonstrate the effectiveness of our approach under limited communication ranges, towards deployments under realistic communication constraints.

Solutions to address the periodic review inventory control problem with nonstationary random demand, lost sales, and stochastic vendor lead times typically involve making strong assumptions on the dynamics for either approximation or simulation, and applying methods such as optimization, dynamic programming, or reinforcement learning. Therefore, it is important to analyze and evaluate any inventory control policy, in particular to see if there is room for improvement. We introduce the concept of an equilibrium policy, a desirable property of a policy that intuitively means that, in hindsight, changing only a small fraction of actions does not result in materially more reward. We provide a light-weight contextual bandit-based algorithm to evaluate and occasionally tweak policies, and show that this method achieves favorable guarantees, both theoretically and in empirical studies.