We propose C-LAIfO, a computationally efficient algorithm designed for imitation learning from videos, even in the presence of visual mismatch between agent and expert domains. We analyze the problem of imitation from expert videos with visual discrepancies, and introduce a solution for robust latent space estimation using contrastive learning and data augmentation. Provided a visually robust latent space, our algorithm performs imitation entirely within this space using off-policy adversarial imitation learning. We conduct a thorough ablation study to justify our design choices and test C-LAIfO on high-dimensional continuous robotic tasks. Additionally, we demonstrate how C-LAIfO can be combined with other reward signals to facilitate learning on a set of challenging hand manipulation tasks with sparse rewards. Our experiments show improved performance compared to baseline methods, highlighting the effectiveness and versatility of C-LAIfO. To ensure reproducibility, we provide open access to our code.

The realm of autonomous driving research has witnessed remarkable progress, with simulation technologies [1][2][3][4] reaching unprecedented levels of realism and the burgeoning availability of real-world driving datasets [5][6][7][8]. Despite these advancements, data-driven planning continues to confront a formidable obstacle: the infinite state space and extensive data distribution characteristic of real-world driving. Imitation learning approaches encounter hurdles [9][10] when presented with scenarios that deviate from the training distribution, exemplified by rare events like emergency braking for unforeseen obstacles. Similarly, these methods grapple with long-tail distribution phenomena, such as navigating through unexpected weather conditions or handling the erratic movements of a jaywalking pedestrian. On the other hand, reinforcement learning (RL) strategies aim to cultivate policies through reward-based learning. RL has difficulty bridging the sim-real gap and sampling efficiency [11].

The critical inquiry pervading the realm of Philosophy, and perhaps extending its influence across all Humanities disciplines, revolves around the intricacies of morality and normativity. Surprisingly, in recent years, this thematic thread has woven its way into an unexpected domain, one not conventionally associated with pondering "what ought to be": the field of artificial intelligence (AI) research. Central to morality and AI, we find "alignment", a problem related to the challenges of expressing human goals and values in a manner that artificial systems can follow without leading to unwanted adversarial effects. More explicitly and with our current paradigm of AI development in mind, we can think of alignment as teaching human values to non-anthropomorphic entities trained through opaque, gradient-based learning techniques. This work addresses alignment as a technical-philosophical problem that requires solid philosophical foundations and practical implementations that bring normative theory to AI system development. To accomplish this, we propose two sets of necessary and sufficient conditions that, we argue, should be considered in any alignment process. While necessary conditions serve as metaphysical and metaethical roots that pertain to the permissibility of alignment, sufficient conditions establish a blueprint for aligning AI systems under a learning-based paradigm. After laying such foundations, we present implementations of this approach by using state-of-the-art techniques and methods for aligning general-purpose language systems. We call this framework Dynamic Normativity. Its central thesis is that any alignment process under a learning paradigm that cannot fulfill its necessary and sufficient conditions will fail in producing aligned systems.

In 2021, Adam Zsolt Wagner proposed an approach to disprove conjectures in graph theory using Reinforcement Learning (RL). Wagner's idea can be framed as follows: consider a conjecture, such as a certain quantity f(G) < 0 for every graph G; one can then play a single-player graph-building game, where at each turn the player decides whether to add an edge or not. The game ends when all edges have been considered, resulting in a certain graph G_T, and f(G_T) is the final score of the game; RL is then used to maximize this score. This brilliant idea is as simple as innovative, and it lends itself to systematic generalization. Several different single-player graph-building games can be employed, along with various RL algorithms. Moreover, RL maximizes the cumulative reward, allowing for step-by-step rewards instead of a single final score, provided the final cumulative reward represents the quantity of interest f(G_T). In this paper, we discuss these and various other choices that can be significant in Wagner's framework. As a contribution to this systematization, we present four distinct single-player graph-building games. Each game employs both a step-by-step reward system and a single final score. We also propose a principled approach to select the most suitable neural network architecture for any given conjecture, and introduce a new dataset of graphs labeled with their Laplacian spectra. Furthermore, we provide a counterexample for a conjecture regarding the sum of the matching number and the spectral radius, which is simpler than the example provided in Wagner's original paper. The games have been implemented as environments in the Gymnasium framework, and along with the dataset, are available as open-source supplementary materials.

Fast adaptation to new tasks is extremely important for embodied agents in the real world. Meta-reinforcement learning (meta-RL) has emerged as an effective method to enable fast adaptation in unknown environments. Compared to on-policy meta-RL algorithms, off-policy algorithms rely heavily on efficient data sampling strategies to extract and represent the historical trajectories. However, little is known about how different data sampling methods impact the ability of meta-RL agents to represent unknown environments. Here, we investigate the impact of data sampling strategies on the exploration and adaptability of meta-RL agents. Specifically, we conducted experiments with two types of off-policy meta-RL algorithms based on Thompson sampling and Bayes-optimality theories in continuous control tasks within the MuJoCo environment and sparse reward navigation tasks. Our analysis revealed the long-memory and short-memory sequence sampling strategies affect the representation and adaptive capabilities of meta-RL agents. We found that the algorithm based on Bayes-optimality theory exhibited more robust and better adaptability than the algorithm based on Thompson sampling, highlighting the importance of appropriate data sampling strategies for the agent's representation of an unknown environment, especially in the case of sparse rewards.

Quadcopters have been studied for decades thanks to their maneuverability and capability of operating in a variety of circumstances. However, quadcopters suffer from dynamical nonlinearity, actuator saturation, as well as sensor noise that make it challenging and time consuming to obtain accurate dynamic models and achieve satisfactory control performance. Fortunately, deep reinforcement learning came and has shown significant potential in system modelling and control of autonomous multirotor aerial vehicles, with recent advancements in deployment, performance enhancement, and generalization. In this paper, an end-to-end deep reinforcement learning-based controller for quadcopters is proposed that is secure for real-world implementation, data-efficient, and free of human gain adjustments. First, a novel actor-critic-based architecture is designed to map the robot states directly to the motor outputs. Then, a quadcopter dynamics-based simulator was devised to facilitate the training of the controller policy. Finally, the trained policy is deployed on a real Crazyflie nano quadrotor platform, without any additional fine-tuning process. Experimental results show that the quadcopter exhibits satisfactory performance as it tracks a given complicated trajectory, which demonstrates the effectiveness and feasibility of the proposed method and signifies its capability in filling the simulation-to-reality gap.

The recency heuristic in reinforcement learning is the assumption that stimuli that occurred closer in time to an acquired reward should be more heavily reinforced. The recency heuristic is one of the key assumptions made by TD($\lambda$), which reinforces recent experiences according to an exponentially decaying weighting. In fact, all other widely used return estimators for TD learning, such as $n$-step returns, satisfy a weaker (i.e., non-monotonic) recency heuristic. Why is the recency heuristic effective for temporal credit assignment? What happens when credit is assigned in a way that violates this heuristic? In this paper, we analyze the specific mathematical implications of adopting the recency heuristic in TD learning. We prove that any return estimator satisfying this heuristic: 1) is guaranteed to converge to the correct value function, 2) has a relatively fast contraction rate, and 3) has a long window of effective credit assignment, yet bounded worst-case variance. We also give a counterexample where on-policy, tabular TD methods violating the recency heuristic diverge. Our results offer some of the first theoretical evidence that credit assignment based on the recency heuristic facilitates learning.

Purpose: Autonomous navigation of catheters and guidewires can enhance endovascular surgery safety and efficacy, reducing procedure times and operator radiation exposure. Integrating tele-operated robotics could widen access to time-sensitive emergency procedures like mechanical thrombectomy (MT). Reinforcement learning (RL) shows potential in endovascular navigation, yet its application encounters challenges without a reward signal. This study explores the viability of autonomous navigation in MT vasculature using inverse RL (IRL) to leverage expert demonstrations. Methods: This study established a simulation-based training and evaluation environment for MT navigation. We used IRL to infer reward functions from expert behaviour when navigating a guidewire and catheter. We utilized soft actor-critic to train models with various reward functions and compared their performance in silico. Results: We demonstrated feasibility of navigation using IRL. When evaluating single versus dual device (i.e. guidewire versus catheter and guidewire) tracking, both methods achieved high success rates of 95% and 96%, respectively. Dual-tracking, however, utilized both devices mimicking an expert. A success rate of 100% and procedure time of 22.6 s were obtained when training with a reward function obtained through reward shaping. This outperformed a dense reward function (96%, 24.9 s) and an IRL-derived reward function (48%, 59.2 s). Conclusions: We have contributed to the advancement of autonomous endovascular intervention navigation, particularly MT, by employing IRL. The results underscore the potential of using reward shaping to train models, offering a promising avenue for enhancing the accessibility and precision of MT. We envisage that future research can extend our methodology to diverse anatomical structures to enhance generalizability.

It is widely recognised that communication is key to successful interaction. Humans communicate with each other through both explicit (direct, deliberate communication over an established channel with clear intent to reach a defined recipient [1]) and implicit channels. Implicit communication is a subtle, indirect mode of conveying information, often relying on context, nonverbal cues, and shared understanding between communicators to convey meaning without explicit verbalization [2]. Implicit communication is particularly crucial for human-robot interaction as it enhances a robot's ability to proactively understand and respond to human needs, emotions, and intentions, thereby facilitating more natural and effective communication and collaboration between humans and robots. Unfortunately, most current human-robot interaction (HRI) studies focusing on implicit communication explicitly model the intention of human participants [3] [4], or rely on existing intention knowledge [5].

In offline Imitation Learning (IL), one of the main challenges is the \textit{covariate shift} between the expert observations and the actual distribution encountered by the agent, because it is difficult to determine what action an agent should take when outside the state distribution of the expert demonstrations. Recently, the model-free solutions introduce the supplementary data and identify the latent expert-similar samples to augment the reliable samples during learning. Model-based solutions build forward dynamic models with conservatism quantification and then generate additional trajectories in the neighborhood of expert demonstrations. However, without reward supervision, these methods are often over-conservative in the out-of-expert-support regions, because only in states close to expert-observed states can there be a preferred action enabling policy optimization. To encourage more exploration on expert-unobserved states, we propose a novel model-based framework, called offline Imitation Learning with Self-paced Reverse Augmentation (SRA). Specifically, we build a reverse dynamic model from the offline demonstrations, which can efficiently generate trajectories leading to the expert-observed states in a self-paced style. Then, we use the subsequent reinforcement learning method to learn from the augmented trajectories and transit from expert-unobserved states to expert-observed states. This framework not only explores the expert-unobserved states but also guides maximizing long-term returns on these states, ultimately enabling generalization beyond the expert data. Empirical results show that our proposal could effectively mitigate the covariate shift and achieve the state-of-the-art performance on the offline imitation learning benchmarks. Project website: \url{https://www.lamda.nju.edu.cn/shaojj/KDD24_SRA/}.