Model-based evaluation metrics (e.g., CLIPScore and GPTScore) have demonstrated decent correlations with human judgments in various language generation tasks. However, their impact on fairness remains largely unexplored. It is widely recognized that pretrained models can inadvertently encode societal biases, thus employing these models for evaluation purposes may inadvertently perpetuate and amplify biases. For example, an evaluation metric may favor the caption "a woman is calculating an account book" over "a man is calculating an account book," even if the image only shows male accountants. In this paper, we conduct a systematic study of gender biases in model-based automatic evaluation metrics for image captioning tasks. We start by curating a dataset comprising profession, activity, and object concepts associated with stereotypical gender associations. Then, we demonstrate the negative consequences of using these biased metrics, including the inability to differentiate between biased and unbiased generations, as well as the propagation of biases to generation models through reinforcement learning. Finally, we present a simple and effective way to mitigate the metric bias without hurting the correlations with human judgments. Our dataset and framework lay the foundation for understanding the potential harm of model-based evaluation metrics, and facilitate future works to develop more inclusive evaluation metrics.

This paper introduces a novel pipeline for summarising timelines of events reported by multiple news sources. Transformer-based models for abstractive summarisation generate coherent and concise summaries of long documents but can fail to outperform established extractive methods on specialised tasks such as timeline summarisation (TLS). While extractive summaries are more faithful to their sources, they may be less readable and contain redundant or unnecessary information. This paper proposes a preference-based reinforcement learning (PBRL) method for adapting pretrained abstractive summarisers to TLS, which can overcome the drawbacks of extractive timeline summaries. We define a compound reward function that learns from keywords of interest and pairwise preference labels, which we use to fine-tune a pretrained abstractive summariser via offline reinforcement learning. We carry out both automated and human evaluation on three datasets, finding that our method outperforms a comparable extractive TLS method on two of the three benchmark datasets, and participants prefer our method's summaries to those of both the extractive TLS method and the pretrained abstractive model. The method does not require expensive reference summaries and needs only a small number of preferences to align the generated summaries with human preferences.

Consider learning an imitation policy on the basis of demonstrated behavior from multiple environments, with an eye towards deployment in an unseen environment. Since the observable features from each setting may be different, directly learning individual policies as mappings from features to actions is prone to spurious correlations -- and may not generalize well. However, the expert's policy is often a function of a shared latent structure underlying those observable features that is invariant across settings. By leveraging data from multiple environments, we propose Invariant Causal Imitation Learning (ICIL), a novel technique in which we learn a feature representation that is invariant across domains, on the basis of which we learn an imitation policy that matches expert behavior. To cope with transition dynamics mismatch, ICIL learns a shared representation of causal features (for all training environments), that is disentangled from the specific representations of noise variables (for each of those environments). Moreover, to ensure that the learned policy matches the observation distribution of the expert's policy, ICIL estimates the energy of the expert's observations and uses a regularization term that minimizes the imitator policy's next state energy. Experimentally, we compare our methods against several benchmarks in control and healthcare tasks and show its effectiveness in learning imitation policies capable of generalizing to unseen environments.

Consider learning a generative model for time-series data. The sequential setting poses a unique challenge: Not only should the generator capture the conditional dynamics of (stepwise) transitions, but its open-loop rollouts should also preserve the joint distribution of (multi-step) trajectories. On one hand, autoregressive models trained by MLE allow learning and computing explicit transition distributions, but suffer from compounding error during rollouts. On the other hand, adversarial models based on GAN training alleviate such exposure bias, but transitions are implicit and hard to assess. In this work, we study a generative framework that seeks to combine the strengths of both: Motivated by a moment-matching objective to mitigate compounding error, we optimize a local (but forward-looking) transition policy, where the reinforcement signal is provided by a global (but stepwise-decomposable) energy model trained by contrastive estimation. At training, the two components are learned cooperatively, avoiding the instabilities typical of adversarial objectives. At inference, the learned policy serves as the generator for iterative sampling, and the learned energy serves as a trajectory-level measure for evaluating sample quality. By expressly training a policy to imitate sequential behavior of time-series features in a dataset, this approach embodies "generation by imitation". Theoretically, we illustrate the correctness of this formulation and the consistency of the algorithm. Empirically, we evaluate its ability to generate predictively useful samples from real-world datasets, verifying that it performs at the standard of existing benchmarks.

The goal of this paper is to demonstrate that common noise may serve as an exploration noise for learning the solution of a mean field game. This concept is here exemplified through a toy linear-quadratic model, for which a suitable form of common noise has already been proven to restore existence and uniqueness. We here go one step further and prove that the same form of common noise may force the convergence of the learning algorithm called `fictitious play', and this without any further potential or monotone structure. Several numerical examples are provided in order to support our theoretical analysis.

Urban search and rescue missions require rapid first response to minimize loss of life and damage. Often, such efforts are assisted by humanitarian robots which need to handle dynamic operational conditions such as uneven and rough terrains, especially during mass casualty incidents like an earthquake. Quadruped robots, owing to their versatile design, have the potential to assist in such scenarios. However, control of quadruped robots in dynamic and rough terrain environments is a challenging problem due to the many degrees of freedom of these robots. Current locomotion controllers for quadrupeds are limited in their ability to produce multiple adaptive gaits, solve tasks in a time and resource-efficient manner, and require tedious training and manual tuning procedures. To address these challenges, we propose MTAC: a multi-gait terrain-adaptive controller, which utilizes a Hierarchical reinforcement learning (HRL) approach while being time and memory-efficient. We show that our proposed method scales well to a diverse range of environments with similar compute times as state-of-the-art methods. Our method showed greater than 75% on most tasks, outperforming previous work on the majority of test cases.

In robotics and artificial intelligence, the integration of tactile processing is becoming increasingly pivotal, especially in learning to execute intricate tasks like alignment and insertion. However, existing works focusing on tactile methods for insertion tasks predominantly rely on robot teleoperation data and reinforcement learning, which do not utilize the rich insights provided by human's control strategy guided by tactile feedback. For utilizing human sensations, methodologies related to learning from humans predominantly leverage visual feedback, often overlooking the invaluable tactile feedback that humans inherently employ to finish complex manipulations. Addressing this gap, we introduce "MimicTouch", a novel framework that mimics human's tactile-guided control strategy. In this framework, we initially collect multi-modal tactile datasets from human demonstrators, incorporating human tactile-guided control strategies for task completion. The subsequent step involves instructing robots through imitation learning using multi-modal sensor data and retargeted human motions. To further mitigate the embodiment gap between humans and robots, we employ online residual reinforcement learning on the physical robot. Through comprehensive experiments, we validate the safety of MimicTouch in transferring a latent policy learned through imitation learning from human to robot. This ongoing work will pave the way for a broader spectrum of tactile-guided robotic applications.

Deep reinforcement learning (DRL) has led to a wide range of advances in sequential decision-making tasks. However, the complexity of neural network policies makes it difficult to understand and deploy with limited computational resources. Currently, employing compact symbolic expressions as symbolic policies is a promising strategy to obtain simple and interpretable policies. Previous symbolic policy methods usually involve complex training processes and pre-trained neural network policies, which are inefficient and limit the application of symbolic policies. In this paper, we propose an efficient gradient-based learning method named Efficient Symbolic Policy Learning (ESPL) that learns the symbolic policy from scratch in an end-to-end way. We introduce a symbolic network as the search space and employ a path selector to find the compact symbolic policy. By doing so we represent the policy with a differentiable symbolic expression and train it in an off-policy manner which further improves the efficiency. In addition, in contrast with previous symbolic policies which only work in single-task RL because of complexity, we expand ESPL on meta-RL to generate symbolic policies for unseen tasks. Experimentally, we show that our approach generates symbolic policies with higher performance and greatly improves data efficiency for single-task RL. In meta-RL, we demonstrate that compared with neural network policies the proposed symbolic policy achieves higher performance and efficiency and shows the potential to be interpretable.

Federated learning enables a collaborative training and optimization of global models among a group of devices without sharing local data samples. However, the heterogeneity of data in federated learning can lead to unfair representation of the global model across different devices. To address the fairness issue in federated learning, we propose a dynamic q fairness federated learning algorithm with reinforcement learning, called DQFFL. DQFFL aims to mitigate the discrepancies in device aggregation and enhance the fairness of treatment for all groups involved in federated learning. To quantify fairness, DQFFL leverages the performance of the global federated model on each device and incorporates {\alpha}-fairness to transform the preservation of fairness during federated aggregation into the distribution of client weights in the aggregation process. Considering the sensitivity of parameters in measuring fairness, we propose to utilize reinforcement learning for dynamic parameters during aggregation. Experimental results demonstrate that our DQFFL outperforms the state-of-the-art methods in terms of overall performance, fairness and convergence speed.

The development of trustworthy conversational information-seeking systems relies on dialogue models that can generate faithful and accurate responses based on relevant knowledge texts. However, two main challenges hinder this task. Firstly, language models may generate hallucinations due to data biases present in their pretraining corpus. Secondly, knowledge texts often contain redundant and irrelevant information that distracts the model's attention from the relevant text span. Previous works use additional data annotations on the knowledge texts to learn a knowledge identification module in order to bypass irrelevant information, but collecting such high-quality span annotations can be costly. In this work, we leverage reinforcement learning algorithms to overcome the above challenges by introducing a novel reward function. Our reward function combines an accuracy metric and a faithfulness metric to provide a balanced quality judgment of generated responses, which can be used as a cost-effective approximation to a human preference reward model when only a few preference annotations are available. Empirical experiments on two conversational information-seeking datasets demonstrate that our method can compete with other strong supervised learning baselines.