Learning high-quality Q-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that Q-values are indeed underestimated in the latter stage of the RL training process, primarily related to the use of inferior actions from the current policy in Bellman updates as compared to the more optimal action samples in the replay buffer. We hypothesize that this long-neglected phenomenon potentially hinders policy learning and reduces sample efficiency. Our insight to address this issue is to incorporate sufficient exploitation of past successes while maintaining exploration optimism. We propose the Blended Exploitation and Exploration (BEE) operator, a simple yet effective approach that updates Q-value using both historical best-performing actions and the current policy. The instantiations of our method in both model-free and model-based settings outperform state-of-the-art methods in various continuous control tasks and achieve strong performance in failure-prone scenarios and real-world robot tasks.
相關內容
The paper applies reinforcement learning to novel Internet of Thing configurations. Our analysis of inaudible attacks on voice-activated devices confirms the alarming risk factor of 7.6 out of 10, underlining significant security vulnerabilities scored independently by NIST National Vulnerability Database (NVD). Our baseline network model showcases a scenario in which an attacker uses inaudible voice commands to gain unauthorized access to confidential information on a secured laptop. We simulated many attack scenarios on this baseline network model, revealing the potential for mass exploitation of interconnected devices to discover and own privileged information through physical access without adding new hardware or amplifying device skills. Using Microsoft's CyberBattleSim framework, we evaluated six reinforcement learning algorithms and found that Deep-Q learning with exploitation proved optimal, leading to rapid ownership of all nodes in fewer steps. Our findings underscore the critical need for understanding non-conventional networks and new cybersecurity measures in an ever-expanding digital landscape, particularly those characterized by mobile devices, voice activation, and non-linear microphones susceptible to malicious actors operating stealth attacks in the near-ultrasound or inaudible ranges. By 2024, this new attack surface might encompass more digital voice assistants than people on the planet yet offer fewer remedies than conventional patching or firmware fixes since the inaudible attacks arise inherently from the microphone design and digital signal processing.
Machine Learning (ML) models are widely employed to drive many modern data systems. While they are undeniably powerful tools, ML models often demonstrate imbalanced performance and unfair behaviors. The root of this problem often lies in the fact that different subpopulations commonly display divergent trends: as a learning algorithm tries to identify trends in the data, it naturally favors the trends of the majority groups, leading to a model that performs poorly and unfairly for minority populations. Our goal is to improve the fairness and trustworthiness of ML models by applying only non-invasive interventions, i.e., without altering the data or the learning algorithm. We use a simple but key insight: the divergence of trends between different populations, and, consecutively, between a learned model and minority populations, is analogous to data drift, which indicates the poor conformance between parts of the data and the trained model. We explore two strategies (model-splitting and reweighing) to resolve this drift, aiming to improve the overall conformance of models to the underlying data. Both our methods introduce novel ways to employ the recently-proposed data profiling primitive of Conformance Constraints. Our experimental evaluation over 7 real-world datasets shows that both DifFair and ConFair improve the fairness of ML models. We demonstrate scenarios where DifFair has an edge, though ConFair has the greatest practical impact and outperforms other baselines. Moreover, as a model-agnostic technique, ConFair stays robust when used against different models than the ones on which the weights have been learned, which is not the case for other state of the art.
Theoretical guarantees in reinforcement learning (RL) are known to suffer multiplicative blow-up factors with respect to the misspecification error of function approximation. Yet, the nature of such \emph{approximation factors} -- especially their optimal form in a given learning problem -- is poorly understood. In this paper we study this question in linear off-policy value function estimation, where many open questions remain. We study the approximation factor in a broad spectrum of settings, such as with the weighted $L_2$-norm (where the weighting is the offline state distribution), the $L_\infty$ norm, the presence vs. absence of state aliasing, and full vs. partial coverage of the state space. We establish the optimal asymptotic approximation factors (up to constants) for all of these settings. In particular, our bounds identify two instance-dependent factors for the $L_2(\mu)$ norm and only one for the $L_\infty$ norm, which are shown to dictate the hardness of off-policy evaluation under misspecification.
In this study, we propose an improvement to the direct mating method, a constraint handling approach for multi-objective evolutionary algorithms, by hybridizing it with local mating. Local mating selects another parent from the feasible solution space around the initially selected parent. The direct mating method selects the other parent along the optimal direction in the objective space after the first parent is selected, even if it is infeasible. It shows better exploration performance for constraint optimization problems with coupling NSGA-II, but requires several individuals along the optimal direction. Due to the lack of better solutions dominated by the optimal direction from the first parent, direct mating becomes difficult as the generation proceeds. To address this issue, we propose a hybrid method that uses local mating to select another parent from the neighborhood of the first selected parent, maintaining diversity around good solutions and helping the direct mating process. We evaluate the proposed method on three mathematical problems with unique Pareto fronts and two real-world applications. We use the generation histories of the averages and standard deviations of the hypervolumes as the performance evaluation criteria. Our investigation results show that the proposed method can solve constraint multi-objective problems better than existing methods while maintaining high diversity.
ChatGPT, as a versatile large language model, has demonstrated remarkable potential in addressing inquiries across various domains. Its ability to analyze, comprehend, and synthesize information from both online sources and user inputs has garnered significant attention. Previous research has explored ChatGPT's competence in code generation and code reviews. In this paper, we delve into ChatGPT's capabilities in security-oriented program analysis, focusing on perspectives from both attackers and security analysts. We present a case study involving several security-oriented program analysis tasks while deliberately introducing challenges to assess ChatGPT's responses. Through an examination of the quality of answers provided by ChatGPT, we gain a clearer understanding of its strengths and limitations in the realm of security-oriented program analysis.
Machine learning models are often personalized with categorical attributes that are protected, sensitive, self-reported, or costly to acquire. In this work, we show models that are personalized with group attributes can reduce performance at a group level. We propose formal conditions to ensure the "fair use" of group attributes in prediction tasks by training one additional model -- i.e., collective preference guarantees to ensure that each group who provides personal data will receive a tailored gain in performance in return. We present sufficient conditions to ensure fair use in empirical risk minimization and characterize failure modes that lead to fair use violations due to standard practices in model development and deployment. We present a comprehensive empirical study of fair use in clinical prediction tasks. Our results demonstrate the prevalence of fair use violations in practice and illustrate simple interventions to mitigate their harm.
Imitation learning is a primary approach to improve the efficiency of reinforcement learning by exploiting the expert demonstrations. However, in many real scenarios, obtaining expert demonstrations could be extremely expensive or even impossible. To overcome this challenge, in this paper, we propose a novel learning framework called Co-Imitation Learning (CoIL) to exploit the past good experiences of the agents themselves without expert demonstration. Specifically, we train two different agents via letting each of them alternately explore the environment and exploit the peer agent's experience. While the experiences could be valuable or misleading, we propose to estimate the potential utility of each piece of experience with the expected gain of the value function. Thus the agents can selectively imitate from each other by emphasizing the more useful experiences while filtering out noisy ones. Experimental results on various tasks show significant superiority of the proposed Co-Imitation Learning framework, validating that the agents can benefit from each other without external supervision.
In this paper, we present a model describing the collective motion of birds. We explore the dynamic relationship between followers and leaders, wherein a select few agents, known as leaders, can initiate spontaneous changes in direction without being influenced by external factors like predators. Starting at the microscopic level, we develop a kinetic model that characterizes the behaviour of large crowds with transient leadership. One significant challenge lies in managing topological interactions, as identifying nearest neighbors in extensive systems can be computationally expensive. To address this, we propose a novel stochastic particle method to simulate the mesoscopic dynamics and reduce the computational cost of identifying closer agents from quadratic to logarithmic complexity using a $k$-nearest neighbours search algorithm with a binary tree. Lastly, we conduct various numerical experiments for different scenarios to validate the algorithm's effectiveness and investigate collective dynamics in both two and three dimensions.
In this paper, we present an information-theoretic perspective to group fairness trade-offs in federated learning (FL) with respect to sensitive attributes, such as gender, race, etc. Existing works mostly focus on either \emph{global fairness} (overall disparity of the model across all clients) or \emph{local fairness} (disparity of the model at each individual client), without always considering their trade-offs. There is a lack of understanding of the interplay between global and local fairness in FL, and if and when one implies the other. To address this gap, we leverage a body of work in information theory called partial information decomposition (PID) which first identifies three sources of unfairness in FL, namely, \emph{Unique Disparity}, \emph{Redundant Disparity}, and \emph{Masked Disparity}. Using canonical examples, we demonstrate how these three disparities contribute to global and local fairness. This decomposition helps us derive fundamental limits and trade-offs between global or local fairness, particularly under data heterogeneity, as well as, derive conditions under which one implies the other. We also present experimental results on benchmark datasets to support our theoretical findings. This work offers a more nuanced understanding of the sources of disparity in FL that can inform the use of local disparity mitigation techniques, and their convergence and effectiveness when deployed in practice.
Clustering is a fundamental machine learning task which has been widely studied in the literature. Classic clustering methods follow the assumption that data are represented as features in a vectorized form through various representation learning techniques. As the data become increasingly complicated and complex, the shallow (traditional) clustering methods can no longer handle the high-dimensional data type. With the huge success of deep learning, especially the deep unsupervised learning, many representation learning techniques with deep architectures have been proposed in the past decade. Recently, the concept of Deep Clustering, i.e., jointly optimizing the representation learning and clustering, has been proposed and hence attracted growing attention in the community. Motivated by the tremendous success of deep learning in clustering, one of the most fundamental machine learning tasks, and the large number of recent advances in this direction, in this paper we conduct a comprehensive survey on deep clustering by proposing a new taxonomy of different state-of-the-art approaches. We summarize the essential components of deep clustering and categorize existing methods by the ways they design interactions between deep representation learning and clustering. Moreover, this survey also provides the popular benchmark datasets, evaluation metrics and open-source implementations to clearly illustrate various experimental settings. Last but not least, we discuss the practical applications of deep clustering and suggest challenging topics deserving further investigations as future directions.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191