A key challenge for a reinforcement learning (RL) agent is to incorporate external/expert1 advice in its learning. The desired goals of an algorithm that can shape the learning of an RL agent with external advice include (a) maintaining policy invariance; (b) accelerating the learning of the agent; and (c) learning from arbitrary advice [3]. To address this challenge this paper formulates the problem of incorporating external advice in RL as a multi-armed bandit called shaping-bandits. The reward of each arm of shaping bandits corresponds to the return obtained by following the expert or by following a default RL algorithm learning on the true environment reward.We show that directly applying existing bandit and shaping algorithms that do not reason about the non-stationary nature of the underlying returns can lead to poor results. Thus we propose UCB-PIES (UPIES), Racing-PIES (RPIES), and Lazy PIES (LPIES) three different shaping algorithms built on different assumptions that reason about the long-term consequences of following the expert policy or the default RL algorithm. Our experiments in four different settings show that these proposed algorithms achieve the above-mentioned goals whereas the other algorithms fail to do so.
相關內容
Centralized training is widely utilized in the field of multi-agent reinforcement learning (MARL) to assure the stability of training process. Once a joint policy is obtained, it is critical to design a value function factorization method to extract optimal decentralized policies for the agents, which needs to satisfy the individual-global-max (IGM) principle. While imposing additional limitations on the IGM function class can help to meet the requirement, it comes at the cost of restricting its application to more complex multi-agent environments. In this paper, we propose QFree, a universal value function factorization method for MARL. We start by developing mathematical equivalent conditions of the IGM principle based on the advantage function, which ensures that the principle holds without any compromise, removing the conservatism of conventional methods. We then establish a more expressive mixing network architecture that can fulfill the equivalent factorization. In particular, the novel loss function is developed by considering the equivalent conditions as regularization term during policy evaluation in the MARL algorithm. Finally, the effectiveness of the proposed method is verified in a nonmonotonic matrix game scenario. Moreover, we show that QFree achieves the state-of-the-art performance in a general-purpose complex MARL benchmark environment, Starcraft Multi-Agent Challenge (SMAC).
Personalized federated learning (PFL) aims to harness the collective wisdom of clients' data while building personalized models tailored to individual clients' data distributions. Existing works offer personalization primarily to clients who participate in the FL process, making it hard to encompass new clients who were absent or newly show up. In this paper, we propose FedBasis, a novel PFL framework to tackle such a deficiency. FedBasis learns a set of few shareable ``basis'' models, which can be linearly combined to form personalized models for clients. Specifically for a new client, only a small set of combination coefficients, not the model weights, needs to be learned. This notion makes FedBasis more parameter-efficient, robust, and accurate than competitive PFL baselines, especially in the low data regime, without increasing the inference cost. To demonstrate the effectiveness and applicability of FedBasis, we also present a more practical PFL testbed for image classification, featuring larger data discrepancies across clients in both the image and label spaces as well as more faithful training and test splits.
We propose Probabilistic Warp Consistency, a weakly-supervised learning objective for semantic matching. Our approach directly supervises the dense matching scores predicted by the network, encoded as a conditional probability distribution. We first construct an image triplet by applying a known warp to one of the images in a pair depicting different instances of the same object class. Our probabilistic learning objectives are then derived using the constraints arising from the resulting image triplet. We further account for occlusion and background clutter present in real image pairs by extending our probabilistic output space with a learnable unmatched state. To supervise it, we design an objective between image pairs depicting different object classes. We validate our method by applying it to four recent semantic matching architectures. Our weakly-supervised approach sets a new state-of-the-art on four challenging semantic matching benchmarks. Lastly, we demonstrate that our objective also brings substantial improvements in the strongly-supervised regime, when combined with keypoint annotations.
With continuous advances in deep learning, distributed training is becoming common in GPU clusters. Specifically, for emerging workloads with diverse amounts, ratios, and patterns of communication, we observe that network contention can significantly degrade training throughput. However, widely used scheduling policies often face limitations as they are agnostic to network contention between jobs. In this paper, we present a new approach to mitigate network contention in GPU clusters using reinforcement learning. We formulate GPU cluster scheduling as a reinforcement learning problem and opt to learn a network contention-aware scheduling policy that efficiently captures contention sensitivities and dynamically adapts scheduling decisions through continuous evaluation and improvement. We show that compared to widely used scheduling policies, our approach reduces average job completion time by up to 18.2\% and effectively cuts the tail job completion time by up to 20.7\% while allowing a preferable trade-off between average job completion time and resource utilization.
Federated learning (FL) offers a decentralized training approach for machine learning models, prioritizing data privacy. However, the inherent heterogeneity in FL networks, arising from variations in data distribution, size, and device capabilities, poses challenges in user federation. Recognizing this, Personalized Federated Learning (PFL) emphasizes tailoring learning processes to individual data profiles. In this paper, we address the complexity of clustering users in PFL, especially in dynamic networks, by introducing a dynamic Upper Confidence Bound (dUCB) algorithm inspired by the multi-armed bandit (MAB) approach. The dUCB algorithm ensures that new users can effectively find the best cluster for their data distribution by balancing exploration and exploitation. The performance of our algorithm is evaluated in various cases, showing its effectiveness in handling dynamic federated learning scenarios.
Recent advances in self-supervised learning have highlighted the efficacy of data augmentation in learning data representation from unlabeled data. Training a linear model atop these enhanced representations can yield an adept classifier. Despite the remarkable empirical performance, the underlying mechanisms that enable data augmentation to unravel nonlinear data structures into linearly separable representations remain elusive. This paper seeks to bridge this gap by investigating under what conditions learned representations can linearly separate manifolds when data is drawn from a multi-manifold model. Our investigation reveals that data augmentation offers additional information beyond observed data and can thus improve the information-theoretic optimal rate of linear separation capacity. In particular, we show that self-supervised learning can linearly separate manifolds with a smaller distance than unsupervised learning, underscoring the additional benefits of data augmentation. Our theoretical analysis further underscores that the performance of downstream linear classifiers primarily hinges on the linear separability of data representations rather than the size of the labeled data set, reaffirming the viability of constructing efficient classifiers with limited labeled data amid an expansive unlabeled data set.
Reinforcement learning (RL) has shown great promise for developing dialogue management (DM) agents that are non-myopic, conduct rich conversations, and maximize overall user satisfaction. Despite recent developments in RL and language models (LMs), using RL to power conversational chatbots remains challenging, in part because RL requires online exploration to learn effectively, whereas collecting novel human-bot interactions can be expensive and unsafe. This issue is exacerbated by the combinatorial action spaces facing these algorithms, as most LM agents generate responses at the word level. We develop a variety of RL algorithms, specialized to dialogue planning, that leverage recent Mixture-of-Expert Language Models (MoE-LMs) -- models that capture diverse semantics, generate utterances reflecting different intents, and are amenable for multi-turn DM. By exploiting MoE-LM structure, our methods significantly reduce the size of the action space and improve the efficacy of RL-based DM. We evaluate our methods in open-domain dialogue to demonstrate their effectiveness w.r.t.\ the diversity of intent in generated utterances and overall DM performance.
The remarkable success of deep learning has prompted interest in its application to medical diagnosis. Even tough state-of-the-art deep learning models have achieved human-level accuracy on the classification of different types of medical data, these models are hardly adopted in clinical workflows, mainly due to their lack of interpretability. The black-box-ness of deep learning models has raised the need for devising strategies to explain the decision process of these models, leading to the creation of the topic of eXplainable Artificial Intelligence (XAI). In this context, we provide a thorough survey of XAI applied to medical diagnosis, including visual, textual, and example-based explanation methods. Moreover, this work reviews the existing medical imaging datasets and the existing metrics for evaluating the quality of the explanations . Complementary to most existing surveys, we include a performance comparison among a set of report generation-based methods. Finally, the major challenges in applying XAI to medical imaging are also discussed.
There recently has been a surge of interest in developing a new class of deep learning (DL) architectures that integrate an explicit time dimension as a fundamental building block of learning and representation mechanisms. In turn, many recent results show that topological descriptors of the observed data, encoding information on the shape of the dataset in a topological space at different scales, that is, persistent homology of the data, may contain important complementary information, improving both performance and robustness of DL. As convergence of these two emerging ideas, we propose to enhance DL architectures with the most salient time-conditioned topological information of the data and introduce the concept of zigzag persistence into time-aware graph convolutional networks (GCNs). Zigzag persistence provides a systematic and mathematically rigorous framework to track the most important topological features of the observed data that tend to manifest themselves over time. To integrate the extracted time-conditioned topological descriptors into DL, we develop a new topological summary, zigzag persistence image, and derive its theoretical stability guarantees. We validate the new GCNs with a time-aware zigzag topological layer (Z-GCNETs), in application to traffic forecasting and Ethereum blockchain price prediction. Our results indicate that Z-GCNET outperforms 13 state-of-the-art methods on 4 time series datasets.
This paper aims to mitigate straggler effects in synchronous distributed learning for multi-agent reinforcement learning (MARL) problems. Stragglers arise frequently in a distributed learning system, due to the existence of various system disturbances such as slow-downs or failures of compute nodes and communication bottlenecks. To resolve this issue, we propose a coded distributed learning framework, which speeds up the training of MARL algorithms in the presence of stragglers, while maintaining the same accuracy as the centralized approach. As an illustration, a coded distributed version of the multi-agent deep deterministic policy gradient(MADDPG) algorithm is developed and evaluated. Different coding schemes, including maximum distance separable (MDS)code, random sparse code, replication-based code, and regular low density parity check (LDPC) code are also investigated. Simulations in several multi-robot problems demonstrate the promising performance of the proposed framework.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191