DRDT3: Diffusion-Refined Decision Test-Time Training Model

Decision Transformer (DT), a trajectory modeling method, has shown competitive performance compared to traditional offline reinforcement learning (RL) approaches on various classic control tasks. However, it struggles to learn optimal policies from suboptimal, reward-labeled trajectories. In this study, we explore the use of conditional generative modeling to facilitate trajectory stitching given its high-quality data generation ability. Additionally, recent advancements in Recurrent Neural Networks (RNNs) have shown their linear complexity and competitive sequence modeling performance over Transformers. We leverage the Test-Time Training (TTT) layer, an RNN that updates hidden states during testing, to model trajectories in the form of DT. We introduce a unified framework, called Diffusion-Refined Decision TTT (DRDT3), to achieve performance beyond DT models. Specifically, we propose the Decision TTT (DT3) module, which harnesses the sequence modeling strengths of both self-attention and the TTT layer to capture recent contextual information and make coarse action predictions. We further integrate DT3 with the diffusion model using a unified optimization objective. With experiments on multiple tasks of Gym and AntMaze in the D4RL benchmark, our DT3 model without diffusion refinement demonstrates improved performance over standard DT, while DRDT3 further achieves superior results compared to state-of-the-art conventional offline RL and DT-based methods.

Goal-Conditioned Data Augmentation for Offline Reinforcement Learning

Offline reinforcement learning (RL) enables policy learning from pre-collected offline datasets, relaxing the need to interact directly with the environment. However, limited by the quality of offline datasets, it generally fails to learn well-qualified policies in suboptimal datasets. To address datasets with insufficient optimal demonstrations, we introduce Goal-cOnditioned Data Augmentation (GODA), a novel goal-conditioned diffusion-based method for augmenting samples with higher quality. Leveraging recent advancements in generative modeling, GODA incorporates a novel return-oriented goal condition with various selection mechanisms. Specifically, we introduce a controllable scaling technique to provide enhanced return-based guidance during data sampling. GODA learns a comprehensive distribution representation of the original offline datasets while generating new data with selectively higher-return goals, thereby maximizing the utility of limited optimal demonstrations. Furthermore, we propose a novel adaptive gated conditioning method for processing noised inputs and conditions, enhancing the capture of goal-oriented guidance. We conduct experiments on the D4RL benchmark and real-world challenges, specifically traffic signal control (TSC) tasks, to demonstrate GODA's effectiveness in enhancing data quality and superior performance compared to state-of-the-art data augmentation methods across various offline RL algorithms.

Robot Policy Learning with Temporal Optimal Transport Reward

Reward specification is one of the most tricky problems in Reinforcement Learning, which usually requires tedious hand engineering in practice. One promising approach to tackle this challenge is to adopt existing expert video demonstrations for policy learning. Some recent work investigates how to learn robot policies from only a single/few expert video demonstrations. For example, reward labeling via Optimal Transport (OT) has been shown to be an effective strategy to generate a proxy reward by measuring the alignment between the robot trajectory and the expert demonstrations. However, previous work mostly overlooks that the OT reward is invariant to temporal order information, which could bring extra noise to the reward signal. To address this issue, in this paper, we introduce the Temporal Optimal Transport (TemporalOT) reward to incorporate temporal order information for learning a more accurate OT-based proxy reward. Extensive experiments on the Meta-world benchmark tasks validate the efficacy of the proposed method. Code is available at: https://github.com/fuyw/TemporalOT

An Online Self-learning Graph-based Lateral Controller for Self-Driving Cars

The hype around self-driving cars has been growing over the past years and has sparked much research. Several modules in self-driving cars are thoroughly investigated to ensure safety, comfort, and efficiency, among which the controller is crucial. The controller module can be categorized into longitudinal and lateral controllers in which the task of the former is to follow the reference velocity, and the latter is to reduce the lateral displacement error from the reference path. Generally, a tuned controller is not sufficient to perform in all environments. Thus, a controller that can adapt to changing conditions is necessary for autonomous driving. Furthermore, these controllers often depend on vehicle models that also need to adapt over time due to varying environments. This paper uses graphs to present novel techniques to learn the vehicle model and the lateral controller online. First, a heterogeneous graph is presented depicting the current states of and inputs to the vehicle. The vehicle model is then learned online using known physical constraints in conjunction with the processing of the graph through a Graph Neural Network structure. Next, another heterogeneous graph - depicting the transition from current to desired states - is processed through another Graph Neural Network structure to generate the steering command on the fly. Finally, the performance of this self-learning model-based lateral controller is evaluated and shown to be satisfactory on an open-source autonomous driving platform called CARLA.

FuRL: Visual-Language Models as Fuzzy Rewards for Reinforcement Learning

In this work, we investigate how to leverage pre-trained visual-language models (VLM) for online Reinforcement Learning (RL). In particular, we focus on sparse reward tasks with pre-defined textual task descriptions. We first identify the problem of reward misalignment when applying VLM as a reward in RL tasks. To address this issue, we introduce a lightweight fine-tuning method, named Fuzzy VLM reward-aided RL (FuRL), based on reward alignment and relay RL. Specifically, we enhance the performance of SAC/DrQ baseline agents on sparse reward tasks by fine-tuning VLM representations and using relay RL to avoid local minima. Extensive experiments on the Meta-world benchmark tasks demonstrate the efficacy of the proposed method. Code is available at: {\footnotesize\url{https://github.com/fuyw/FuRL}}.

An Online Spatial-Temporal Graph Trajectory Planner for Autonomous Vehicles

The autonomous driving industry is expected to grow by over 20 times in the coming decade and, thus, motivate researchers to delve into it. The primary focus of their research is to ensure safety, comfort, and efficiency. An autonomous vehicle has several modules responsible for one or more of the aforementioned items. Among these modules, the trajectory planner plays a pivotal role in the safety of the vehicle and the comfort of its passengers. The module is also responsible for respecting kinematic constraints and any applicable road constraints. In this paper, a novel online spatial-temporal graph trajectory planner is introduced to generate safe and comfortable trajectories. First, a spatial-temporal graph is constructed using the autonomous vehicle, its surrounding vehicles, and virtual nodes along the road with respect to the vehicle itself. Next, the graph is forwarded into a sequential network to obtain the desired states. To support the planner, a simple behavioral layer is also presented that determines kinematic constraints for the planner. Furthermore, a novel potential function is also proposed to train the network. Finally, the proposed planner is tested on three different complex driving tasks, and the performance is compared with two frequently used methods. The results show that the proposed planner generates safe and feasible trajectories while achieving similar or longer distances in the forward direction and comparable comfort ride.

Traffic Signal Control Using Lightweight Transformers: An Offline-to-Online RL Approach

Efficient traffic signal control is critical for reducing traffic congestion and improving overall transportation efficiency. The dynamic nature of traffic flow has prompted researchers to explore Reinforcement Learning (RL) for traffic signal control (TSC). Compared with traditional methods, RL-based solutions have shown preferable performance. However, the application of RL-based traffic signal controllers in the real world is limited by the low sample efficiency and high computational requirements of these solutions. In this work, we propose DTLight, a simple yet powerful lightweight Decision Transformer-based TSC method that can learn policy from easily accessible offline datasets. DTLight novelly leverages knowledge distillation to learn a lightweight controller from a well-trained larger teacher model to reduce implementation computation. Additionally, it integrates adapter modules to mitigate the expenses associated with fine-tuning, which makes DTLight practical for online adaptation with minimal computation and only a few fine-tuning steps during real deployment. Moreover, DTLight is further enhanced to be more applicable to real-world TSC problems. Extensive experiments on synthetic and real-world scenarios show that DTLight pre-trained purely on offline datasets can outperform state-of-the-art online RL-based methods in most scenarios. Experiment results also show that online fine-tuning further improves the performance of DTLight by up to 42.6% over the best online RL baseline methods. In this work, we also introduce Datasets specifically designed for TSC with offline RL (referred to as DTRL). Our datasets and code are publicly available.

Anomaly Detection with Ensemble of Encoder and Decoder

Hacking and false data injection from adversaries can threaten power grids' everyday operations and cause significant economic loss. Anomaly detection in power grids aims to detect and discriminate anomalies caused by cyber attacks against the power system, which is essential for keeping power grids working correctly and efficiently. Different methods have been applied for anomaly detection, such as statistical methods and machine learning-based methods. Usually, machine learning-based methods need to model the normal data distribution. In this work, we propose a novel anomaly detection method by modeling the data distribution of normal samples via multiple encoders and decoders. Specifically, the proposed method maps input samples into a latent space and then reconstructs output samples from latent vectors. The extra encoder finally maps reconstructed samples to latent representations. During the training phase, we optimize parameters by minimizing the reconstruction loss and encoding loss. Training samples are re-weighted to focus more on missed correlations between features of normal data. Furthermore, we employ the long short-term memory model as encoders and decoders to test its effectiveness. We also investigate a meta-learning-based framework for hyper-parameter tuning of our approach. Experiment results on network intrusion and power system datasets demonstrate the effectiveness of our proposed method, where our models consistently outperform all baselines.

Adaptive Aggregation for Safety-Critical Control

Safety has been recognized as the central obstacle to preventing the use of reinforcement learning (RL) for real-world applications. Different methods have been developed to deal with safety concerns in RL. However, learning reliable RL-based solutions usually require a large number of interactions with the environment. Likewise, how to improve the learning efficiency, specifically, how to utilize transfer learning for safe reinforcement learning, has not been well studied. In this work, we propose an adaptive aggregation framework for safety-critical control. Our method comprises two key techniques: 1) we learn to transfer the safety knowledge by aggregating the multiple source tasks and a target task through the attention network; 2) we separate the goal of improving task performance and reducing constraint violations by utilizing a safeguard. Experiment results demonstrate that our algorithm can achieve fewer safety violations while showing better data efficiency compared with several baselines.

MetaEMS: A Meta Reinforcement Learning-based Control Framework for Building Energy Management System

The building sector has been recognized as one of the primary sectors for worldwide energy consumption. Improving the energy efficiency of the building sector can help reduce the operation cost and reduce the greenhouse gas emission. The energy management system (EMS) can monitor and control the operations of built-in appliances in buildings, so an efficient EMS is of crucial importance to improve the building operation efficiency and maintain safe operations. With the growing penetration of renewable energy and electrical appliances, increasing attention has been paid to the development of intelligent building EMS. Recently, reinforcement learning (RL) has been applied for building EMS and has shown promising potential. However, most of the current RL-based EMS solutions would need a large amount of data to learn a reliable control policy, which limits the applicability of these solutions in the real world. In this work, we propose MetaEMS, which can help achieve better energy management performance with the benefits of RL and meta-learning. Experiment results showcase that our proposed MetaEMS can adapt faster to environment changes and perform better in most situations compared with other baselines.