亚洲男人的天堂2018av,欧美草比,久久久久久免费视频精选,国色天香在线看免费,久久久久亚洲av成人片仓井空

Graph rewriting is a popular tool for the optimisation and modification of graph expressions in domains such as compilers, machine learning and quantum computing. The underlying data structures are often port graphs - graphs with labels at edge endpoints. These port labels greatly simplify pattern matching. A pre-requisite for graph rewriting is the ability to find subgraphs of the input that match known graph identities: the pattern matching problem. We propose a new solution to pattern matching in port graphs. Its novelty lies in the use of a pre-computed data structure that makes the pattern matching runtime complexity independent of the number of patterns. The runtime is bound by the maximum width $w$ and depth $d$ of the patterns, as well as the input graph size $|G|$ as $O(|G| \cdot c^w / w^{1/2} \cdot d)$ with $c = 6.75$. This offers a significant advantage over existing solutions for use cases where patterns have low width and the set of patterns is large and fixed ahead of time. In the context of quantum circuits, pattern width can be limited to qubit number. Quantum superoptimisers may use thousands of rewrite rules on circuits with less than 5 qubits, making them an ideal use case. We provide benchmarks showing that our algorithm offers a 20x speedup over current implementations on a dataset of 10'000 real world patterns describing quantum circuits.

Datalog is a popular and widely-used declarative logic programming language. Datalog engines apply many cross-rule optimizations; bugs in them can cause incorrect results. To detect such optimization bugs, we propose an automated testing approach called Incremental Rule Evaluation (IRE), which synergistically tackles the test oracle and test case generation problem. The core idea behind the test oracle is to compare the results of an optimized program and a program without cross-rule optimization; any difference indicates a bug in the Datalog engine. Our core insight is that, for an optimized, incrementally-generated Datalog program, we can evaluate all rules individually by constructing a reference program to disable the optimizations that are performed among multiple rules. Incrementally generating test cases not only allows us to apply the test oracle for every new rule generated-we also can ensure that every newly added rule generates a non-empty result with a given probability and eschew recomputing already-known facts. We implemented IRE as a tool named Deopt, and evaluated Deopt on four mature Datalog engines, namely Souffl\'e, CozoDB, $\mu$Z, and DDlog, and discovered a total of 30 bugs. Of these, 13 were logic bugs, while the remaining were crash and error bugs. Deopt can detect all bugs found by queryFuzz, a state-of-the-art approach. Out of the bugs identified by Deopt, queryFuzz might be unable to detect 5. Our incremental test case generation approach is efficient; for example, for test cases containing 60 rules, our incremental approach can produce 1.17$\times$ (for DDlog) to 31.02$\times$ (for Souffl\'e) as many valid test cases with non-empty results as the naive random method. We believe that the simplicity and the generality of the approach will lead to its wide adoption in practice.

For reinforcement learning on complex stochastic systems, it is desirable to effectively leverage the information from historical samples collected in previous iterations to accelerate policy optimization. Classical experience replay, while effective, treats all observations uniformly, neglecting their relative importance. To address this limitation, we introduce a novel Variance Reduction Experience Replay (VRER) framework, enabling the selective reuse of relevant samples to improve policy gradient estimation. VRER, as an adaptable method that can seamlessly integrate with different policy optimization algorithms, forms the foundation of our sample-efficient off-policy algorithm known as Policy Optimization with VRER (PG-VRER). Furthermore, the lack of a rigorous theoretical understanding of the experience replay method in the literature motivates us to introduce a novel theoretical framework that accounts for sample dependencies induced by Markovian noise and behavior policy interdependencies. This framework is then employed to analyze the finite-time convergence of our VRER-based policy optimization algorithm, revealing a crucial bias-variance trade-off in policy gradient estimates: the reuse of old experience introduces increased bias while simultaneously reducing gradient variance. Extensive experiments have shown that VRER offers a notable acceleration in learning optimal policies and enhances the performance of state-of-the-art (SOTA) policy optimization approaches.

Various data augmentation techniques have been recently proposed in image-based deep reinforcement learning (DRL). Although they empirically demonstrate the effectiveness of data augmentation for improving sample efficiency or generalization, which technique should be preferred is not always clear. To tackle this question, we analyze existing methods to better understand them and to uncover how they are connected. Notably, by expressing the variance of the Q-targets and that of the empirical actor/critic losses of these methods, we can analyze the effects of their different components and compare them. We furthermore formulate an explanation about how these methods may be affected by choosing different data augmentation transformations in calculating the target Q-values. This analysis suggests recommendations on how to exploit data augmentation in a more principled way. In addition, we include a regularization term called tangent prop, previously proposed in computer vision, but whose adaptation to DRL is novel to the best of our knowledge. We evaluate our proposition and validate our analysis in several domains. Compared to different relevant baselines, we demonstrate that it achieves state-of-the-art performance in most environments and shows higher sample efficiency and better generalization ability in some complex environments.

What are the functionals of the reward that can be computed and optimized exactly in Markov Decision Processes?In the finite-horizon, undiscounted setting, Dynamic Programming (DP) can only handle these operations efficiently for certain classes of statistics. We summarize the characterization of these classes for policy evaluation, and give a new answer for the planning problem. Interestingly, we prove that only generalized means can be optimized exactly, even in the more general framework of Distributional Reinforcement Learning (DistRL).DistRL permits, however, to evaluate other functionals approximately. We provide error bounds on the resulting estimators, and discuss the potential of this approach as well as its limitations.These results contribute to advancing the theory of Markov Decision Processes by examining overall characteristics of the return, and particularly risk-conscious strategies.

We consider hypergraph network design problems where the goal is to construct a hypergraph that satisfies certain connectivity requirements. For graph network design problems where the goal is to construct a graph that satisfies certain connectivity requirements, the number of edges in every feasible solution is at most quadratic in the number of vertices. In contrast, for hypergraph network design problems, we might have feasible solutions in which the number of hyperedges is exponential in the number of vertices. This presents an additional technical challenge in hypergraph network design problems compared to graph network design problems: in order to solve the problem in polynomial time, we first need to show that there exists a feasible solution in which the number of hyperedges is polynomial in the input size. The central theme of this work is to show that certain hypergraph network design problems admit solutions in which the number of hyperedges is polynomial in the number of vertices and moreover, can be solved in strongly polynomial time. Our work improves on the previous fastest pseudo-polynomial run-time for these problems. In addition, we develop strongly polynomial time algorithms that return near-uniform hypergraphs as solutions (i.e., every pair of hyperedges differ in size by at most one). As applications of our results, we derive the first strongly polynomial time algorithms for (i) degree-specified hypergraph connectivity augmentation using hyperedges, (ii) degree-specified hypergraph node-to-area connectivity augmentation using hyperedges, and (iii) degree-constrained mixed-hypergraph connectivity augmentation using hyperedges.

Deep neural networks (DNNs) have succeeded in many different perception tasks, e.g., computer vision, natural language processing, reinforcement learning, etc. The high-performed DNNs heavily rely on intensive resource consumption. For example, training a DNN requires high dynamic memory, a large-scale dataset, and a large number of computations (a long training time); even inference with a DNN also demands a large amount of static storage, computations (a long inference time), and energy. Therefore, state-of-the-art DNNs are often deployed on a cloud server with a large number of super-computers, a high-bandwidth communication bus, a shared storage infrastructure, and a high power supplement. Recently, some new emerging intelligent applications, e.g., AR/VR, mobile assistants, Internet of Things, require us to deploy DNNs on resource-constrained edge devices. Compare to a cloud server, edge devices often have a rather small amount of resources. To deploy DNNs on edge devices, we need to reduce the size of DNNs, i.e., we target a better trade-off between resource consumption and model accuracy. In this dissertation, we studied four edge intelligence scenarios, i.e., Inference on Edge Devices, Adaptation on Edge Devices, Learning on Edge Devices, and Edge-Server Systems, and developed different methodologies to enable deep learning in each scenario. Since current DNNs are often over-parameterized, our goal is to find and reduce the redundancy of the DNNs in each scenario.

In the era of deep learning, modeling for most NLP tasks has converged to several mainstream paradigms. For example, we usually adopt the sequence labeling paradigm to solve a bundle of tasks such as POS-tagging, NER, Chunking, and adopt the classification paradigm to solve tasks like sentiment analysis. With the rapid progress of pre-trained language models, recent years have observed a rising trend of Paradigm Shift, which is solving one NLP task by reformulating it as another one. Paradigm shift has achieved great success on many tasks, becoming a promising way to improve model performance. Moreover, some of these paradigms have shown great potential to unify a large number of NLP tasks, making it possible to build a single model to handle diverse tasks. In this paper, we review such phenomenon of paradigm shifts in recent years, highlighting several paradigms that have the potential to solve different NLP tasks.

Data augmentation, the artificial creation of training data for machine learning by transformations, is a widely studied research field across machine learning disciplines. While it is useful for increasing the generalization capabilities of a model, it can also address many other challenges and problems, from overcoming a limited amount of training data over regularizing the objective to limiting the amount data used to protect privacy. Based on a precise description of the goals and applications of data augmentation (C1) and a taxonomy for existing works (C2), this survey is concerned with data augmentation methods for textual classification and aims to achieve a concise and comprehensive overview for researchers and practitioners (C3). Derived from the taxonomy, we divided more than 100 methods into 12 different groupings and provide state-of-the-art references expounding which methods are highly promising (C4). Finally, research perspectives that may constitute a building block for future work are given (C5).

Deep learning has revolutionized many machine learning tasks in recent years, ranging from image classification and video processing to speech recognition and natural language understanding. The data in these tasks are typically represented in the Euclidean space. However, there is an increasing number of applications where data are generated from non-Euclidean domains and are represented as graphs with complex relationships and interdependency between objects. The complexity of graph data has imposed significant challenges on existing machine learning algorithms. Recently, many studies on extending deep learning approaches for graph data have emerged. In this survey, we provide a comprehensive overview of graph neural networks (GNNs) in data mining and machine learning fields. We propose a new taxonomy to divide the state-of-the-art graph neural networks into different categories. With a focus on graph convolutional networks, we review alternative architectures that have recently been developed; these learning paradigms include graph attention networks, graph autoencoders, graph generative networks, and graph spatial-temporal networks. We further discuss the applications of graph neural networks across various domains and summarize the open source codes and benchmarks of the existing algorithms on different learning tasks. Finally, we propose potential research directions in this fast-growing field.