Fine-tuning foundation models has emerged as a powerful approach for generating objects with specific desired properties. Reinforcement learning (RL) provides an effective framework for this purpose, enabling models to generate outputs that maximize a given reward function. However, in many applications such as text generation and drug discovery, it can be suboptimal to optimize using a single reward signal, as multiple evaluation criteria are often necessary. This paper proposes a novel reinforcement learning-based method for fine-tuning foundation models using multiple reward signals. By employing an iterative fine-tuning strategy across these rewards, our approach generalizes state-of-the-art RL-based methods. We further provide a theoretical analysis that offers insights into the performance of multi-reward RL fine-tuning. Experimental results across diverse domains including text, biological sequence, and small molecule generation, demonstrate the effectiveness of the proposed algorithm compared to state-of-the-art baselines.

Federated learning is renowned for its efficacy in distributed model training, ensuring that users, called clients, retain data privacy by not disclosing their data to the central server that orchestrates collaborations. Most previous work on federated learning assumes that clients possess static batches of training data. However, clients may also need to make real-time predictions on streaming data in non-stationary environments. In such dynamic environments, employing pre-trained models may be inefficient, as they struggle to adapt to the constantly evolving data streams. To address this challenge, clients can fine-tune models online, leveraging their observed data to enhance performance. Despite the potential benefits of client participation in federated online model fine-tuning, existing analyses have not conclusively demonstrated its superiority over local model fine-tuning. To bridge this gap, the present paper develops a novel personalized federated learning algorithm, wherein each client constructs a personalized model by combining a locally fine-tuned model with multiple federated models learned by the server over time. Theoretical analysis and experiments on real datasets corroborate the effectiveness of this approach for real-time predictions and federated model fine-tuning.

Editing biological sequences has extensive applications in synthetic biology and medicine, such as designing regulatory elements for nucleic-acid therapeutics and treating genetic disorders. The primary objective in biological-sequence editing is to determine the optimal modifications to a sequence which augment certain biological properties while adhering to a minimal number of alterations to ensure predictability and potentially support safety. In this paper, we propose GFNSeqEditor, a novel biological-sequence editing algorithm which builds on the recently proposed area of generative flow networks (GFlowNets). Our proposed GFNSeqEditor identifies elements within a starting seed sequence that may compromise a desired biological property. Then, using a learned stochastic policy, the algorithm makes edits at these identified locations, offering diverse modifications for each sequence to enhance the desired property. The number of edits can be regulated through specific hyperparameters. We conducted extensive experiments on a range of real-world datasets and biological applications, and our results underscore the superior performance of our proposed algorithm compared to existing state-of-the-art sequence editing methods.

Online learning with expert advice is widely used in various machine learning tasks. It considers the problem where a learner chooses one from a set of experts to take advice and make a decision. In many learning problems, experts may be related, henceforth the learner can observe the losses associated with a subset of experts that are related to the chosen one. In this context, the relationship among experts can be captured by a feedback graph, which can be used to assist the learner's decision-making. However, in practice, the nominal feedback graph often entails uncertainties, which renders it impossible to reveal the actual relationship among experts. To cope with this challenge, the present work studies various cases of potential uncertainties and develops novel online learning algorithms to deal with uncertainties while making use of the uncertain feedback graph. The proposed algorithms are proved to enjoy sublinear regret under mild conditions. Experiments on real datasets are presented to demonstrate the effectiveness of the novel algorithms.

Multi-kernel learning (MKL) has been widely used in learning problems involving function learning tasks. Compared with single kernel learning approach which relies on a pre-selected kernel, the advantage of MKL is its flexibility results from combining a dictionary of kernels. However, inclusion of irrelevant kernels in the dictionary may deteriorate the accuracy of MKL, and increase the computational complexity. Faced with this challenge, a novel graph-aided framework is developed to select a subset of kernels from the dictionary with the assistance of a graph. Different graph construction and refinement schemes are developed based on incurred losses or kernel similarities to assist the adaptive selection process. Moreover, to cope with the scenario where data may be collected in a sequential fashion, or cannot be stored in batch due to the massive scale, random feature approximation are adopted to enable online function learning. It is proved that our proposed algorithms enjoy sub-linear regret bounds. Experiments on a number of real datasets showcase the advantages of our novel graph-aided algorithms compared to state-of-the-art alternatives.

Online learning with expert advice is widely used in various machine learning tasks. It considers the problem where a learner chooses one from a set of experts to take advice and make a decision. In many learning problems, experts may be related, henceforth the learner can observe the losses associated with a subset of experts that are related to the chosen one. In this context, the relationship among experts can be captured by a feedback graph, which can be used to assist the learner’s decision-making. However, in practice, the nominal feedback graph often entails uncertainties, which renders it impossible to reveal the actual relationship among experts. To cope with this challenge, the present work develops a novel online learning algorithm to deal with uncertainties while making use of the uncertain feedback graph. The proposed algorithm is proved to enjoy sublinear regret under mild conditions. Experiments on real datasets are presented to demonstrate the effectiveness of the novel algorithm.

Multi-kernel learning (MKL) exhibits well-documented performance in online non-linear function approximation. Federated learning enables a group of learners (called clients) to train an MKL model on the data distributed among clients to perform online non-linear function approximation. There are some challenges in online federated MKL that need to be addressed: i) Communication efficiency especially when a large number of kernels are considered ii) Heterogeneous data distribution among clients. The present paper develops an algorithmic framework to enable clients to communicate with the server to send their updates with affordable communication cost while clients employ a large dictionary of kernels. Utilizing random feature (RF) approximation, the present paper proposes scalable online federated MKL algorithm. We prove that using the proposed online federated MKL algorithm, each client enjoys sub-linear regret with respect to the RF approximation of its best kernel in hindsight, which indicates that the proposed algorithm can effectively deal with heterogeneity of the data distributed among clients. Experimental results on real datasets showcase the advantages of the proposed algorithm compared with other online federated kernel learning ones.

To provide effective wireless connectivity, the use of aerial vehicles has been proposed as a promising solution in a wide variety of applications. In some of these applications, however, to achieve desirable performance, knowing the location of users may be a necessity. One promising approach for solving the problem of obtaining user locations is through the locations of some nodes (called anchors) and measuring the distance between connected nodes in the network. In this paper, we propose prominent techniques to improve the performance of the localization using multiple moving aerial anchors (MAA). In using MAAs as anchor nodes, the anchor nodes in our scenario have movement capability. This provides us with more flexibility to enhance the accuracy of the localization with taking into consideration the power consumption of MAAs. In fact, we propose an MAA placement method by which each user can be connected to at least one MAA while MAAs communicate with as small as possible power consumption. Then, we propose a path planning for MAAs by which distances between users and MAAs can be measured in order to estimate the locations of users. Our simulation results show that our proposed localization scheme can provide highly accurate location estimations.

Multi-kernel learning (MKL) exhibits reliable performance in nonlinear function approximation tasks. Instead of using one kernel, it learns the optimal kernel from a pre-selected dictionary of kernels. The selection of the dictionary has crucial impact on both the performance and complexity of MKL. Specifically, inclusion of a large number of irrelevant kernels may impair the accuracy, and increase the complexity of MKL algorithms. To enhance the accuracy, and alleviate the computational burden, the present paper develops a novel scheme which actively chooses relevant kernels. The proposed framework models the pruned kernel combination as feedback collected from a graph, that is refined ’on the fly.’ Leveraging the random feature approximation, we propose an online scalable multi-kernel learning approach with graph feedback, and prove that the proposed algorithm enjoys sublinear regret. Numerical tests on real datasets demonstrate the effectiveness of the novel approach.