Sim-to-real transfer trains RL agents in the simulated environments and then deploys them in the real world. Sim-to-real transfer has been widely used in practice because it is often cheaper, safer and much faster to collect samples in simulation than in the real world. Despite the empirical success of the sim-to-real transfer, its theoretical foundation is much less understood. In this paper, we study the sim-to-real transfer in continuous domain with partial observations, where the simulated environments and real-world environments are modeled by linear quadratic Gaussian (LQG) systems. We show that a popular robust adversarial training algorithm is capable of learning a policy from the simulated environment that is competitive to the optimal policy in the real-world environment. To achieve our results, we design a new algorithm for infinite-horizon average-cost LQGs and establish a regret bound that depends on the intrinsic complexity of the model class. Our algorithm crucially relies on a novel history clipping scheme, which might be of independent interest.

When deploying Reinforcement Learning (RL) agents into a physical system, we must ensure that these agents are well aware of the underlying constraints. In many real-world problems, however, the constraints are often hard to specify mathematically and unknown to the RL agents. To tackle these issues, Inverse Constrained Reinforcement Learning (ICRL) empirically estimates constraints from expert demonstrations. As an emerging research topic, ICRL does not have common benchmarks, and previous works tested algorithms under hand-crafted environments with manually-generated expert demonstrations. In this paper, we construct an ICRL benchmark in the context of RL application domains, including robot control, and autonomous driving. For each environment, we design relevant constraints and train expert agents to generate demonstration data. Besides, unlike existing baselines that learn a deterministic constraint, we propose a variational ICRL method to model a posterior distribution of candidate constraints. We conduct extensive experiments on these algorithms under our benchmark and show how they can facilitate studying important research challenges for ICRL. The benchmark, including the instructions for reproducing ICRL algorithms, is available at https://github.com/Guiliang/ICRL-benchmarks-public.

Human-AI shared control allows human to interact and collaborate with autonomous agents to accomplish control tasks in complex environments. Previous Reinforcement Learning (RL) methods attempted goal-conditioned designs to achieve human-controllable policies at the cost of redesigning the reward function and training paradigm. Inspired by the neuroscience approach to investigate the motor cortex in primates, we develop a simple yet effective frequency-based approach called Policy Dissection to align the intermediate representation of the learned neural controller with the kinematic attributes of the agent behavior. Without modifying the neural controller or retraining the model, the proposed approach can convert a given RL-trained policy into a human-controllable policy. We evaluate the proposed approach on many RL tasks such as autonomous driving and locomotion. The experiments show that human-AI shared control system achieved by Policy Dissection in driving task can substantially improve the performance and safety in unseen traffic scenes. With human in the inference loop, the locomotion robots also exhibit versatile controllable motion skills even though they are only trained to move forward. Our results suggest the promising direction of implementing human-AI shared autonomy through interpreting the learned representation of the autonomous agents. Code and demo videos are available at https://metadriverse.github.io/policydissect.

We consider the problem of explaining the temporal behavior of black-box systems using human-interpretable models. To this end, based on recent research trends, we rely on the fundamental yet interpretable models of deterministic finite automata (DFAs) and linear temporal logic (LTL) formulas. In contrast to most existing works for learning DFAs and LTL formulas, we rely on only positive examples. Our motivation is that negative examples are generally difficult to observe, in particular, from black-box systems. To learn meaningful models from positive examples only, we design algorithms that rely on conciseness and language minimality of models as regularizers. To this end, our algorithms adopt two approaches: a symbolic and a counterexample-guided one. While the symbolic approach exploits an efficient encoding of language minimality as a constraint satisfaction problem, the counterexample-guided one relies on generating suitable negative examples to prune the search. Both the approaches provide us with effective algorithms with theoretical guarantees on the learned models. To assess the effectiveness of our algorithms, we evaluate all of them on synthetic data.

The dynamic and evolutionary nature of service requirements in wireless networks has motivated the telecom industry to consider intelligent self-adapting Reinforcement Learning (RL) agents for controlling the growing portfolio of network services. Infusion of many new types of services is anticipated with future adoption of 6G networks, and sometimes these services will be defined by applications that are external to the network. An RL agent trained for managing the needs of a specific service type may not be ideal for managing a different service type without domain adaptation. We provide a simple heuristic for evaluating a measure of proximity between a new service and existing services, and show that the RL agent of the most proximal service rapidly adapts to the new service type through a well defined process of domain adaptation. Our approach enables a trained source policy to adapt to new situations with changed dynamics without retraining a new policy, thereby achieving significant computing and cost-effectiveness. Such domain adaptation techniques may soon provide a foundation for more generalized RL-based service management under the face of rapidly evolving service types.

A core issue in multi-agent federated reinforcement learning is defining how to aggregate insights from multiple agents. This is commonly done by taking the average of each participating agent's model weights into one common model (FedAvg). We instead propose FedFormer, a novel federation strategy that utilizes Transformer Attention to contextually aggregate embeddings from models originating from different learner agents. In so doing, we attentively weigh the contributions of other agents with respect to the current agent's environment and learned relationships, thus providing a more effective and efficient federation. We evaluate our methods on the Meta-World environment and find that our approach yields significant improvements over FedAvg and non-federated Soft Actor-Critic single-agent methods. Our results compared to Soft Actor-Critic show that FedFormer achieves higher episodic return while still abiding by the privacy constraints of federated learning. Finally, we also demonstrate improvements in effectiveness with increased agent pools across all methods in certain tasks. This is contrasted by FedAvg, which fails to make noticeable improvements when scaled.

The sparse representation of graphs has shown great potential for accelerating the computation of graph applications (e.g., Social Networks, Knowledge Graphs) on traditional computing architectures (CPU, GPU, or TPU). But the exploration of large-scale sparse graph computing on processing-in-memory (PIM) platforms (typically with memristive crossbars) is still in its infancy. To implement the computation or storage of large-scale or batch graphs on memristive crossbars, a natural assumption is that a large-scale crossbar is demanded, but with low utilization. Some recent works question this assumption, to avoid the waste of storage and computational resource, the fixed-size or progressively scheduled ''block partition'' schemes are proposed. However, these methods are coarse-grained or static, and are not effectively sparsity-aware. This work proposes the dynamic sparsity-aware mapping scheme generating method that models the problem with a sequential decision-making model, and optimizes it by reinforcement learning (RL) algorithm (REINFORCE). Our generating model (LSTM, combined with the dynamic-fill scheme) generates remarkable mapping performance on a small-scale graph/matrix data (complete mapping costs 43% area of the original matrix) and two large-scale matrix data (costing 22.5% area on qh882 and 17.1% area on qh1484). Our method may be extended to sparse graph computing on other PIM architectures, not limited to the memristive device-based platforms.

Efficient aerial data collection is important in many remote sensing applications. In large-scale monitoring scenarios, deploying a team of unmanned aerial vehicles (UAVs) offers improved spatial coverage and robustness against individual failures. However, a key challenge is cooperative path planning for the UAVs to efficiently achieve a joint mission goal. We propose a novel multi-agent informative path planning approach based on deep reinforcement learning for adaptive terrain monitoring scenarios using UAV teams. We introduce new network feature representations to effectively learn path planning in a 3D workspace. By leveraging a counterfactual baseline, our approach explicitly addresses credit assignment to learn cooperative behaviour. Our experimental evaluation shows improved planning performance, i.e. maps regions of interest more quickly, with respect to non-counterfactual variants. Results on synthetic and real-world data show that our approach has superior performance compared to state-of-the-art non-learning-based methods, while being transferable to varying team sizes and communication constraints.

I was halfway through a master's in Computer Science when my vision changed. I was working as a data scientist during my summer off from school, and I had friends who said things like "I'm at a local maximum" in normal conversation. Wherever I looked, suboptimality began standing out as though highlighted in the angry red of a programming error. There was the athletic-looking youth whose slow amble blocked the walking path so no one else could pass--suboptimal. There was the friend I ran errands with who planned stops in an inefficient order, so the driving took three Nicki Minaj songs longer than it needed to--suboptimal.

Prompting is a promising approach to solving NLP problems with pre-trained language models (LMs) such as GPTs and BERT. Unlike conventional fine-tuning that updates the massive LM parameters for each downstream task, prompting concatenates inputs with additional text to steer the LM towards producing the desired outputs. A key question is finding optimal prompts to improve the LM's performance on various tasks with few training examples. Reinforcement Learning (RL) for prompt optimization challenges learning efficiency as the large black-box language model navigates a complex environment involving multiple transitions before computing rewards. This complexity makes it challenging to learn from the unstable reward signals.