Verified compositional compilation (VCC) is a notion of modular verification of compilers that supports compilation of heterogeneous programs. The key to achieve VCC is to design a semantic interface that enables composition of correctness theorems for compiling individual modules. Most of the existing techniques for VCC fix a semantic interface from the very beginning and force it down to every single compiler pass. This requires significant changes to the existing framework and makes it difficult to understand the relationship between conditions enforced by the semantic interface and the actual requirements of compiler passes. A different approach is to design appropriate semantic interfaces for individual compiler passes and combine them into a unified interface which faithfully reflects the requirements of underlying compiler passes. However, this requires vertically composable simulation relations, which were traditionally considered very difficult to construct even with extensive changes to compiler verification frameworks. We propose a solution to construction of unified semantic interfaces for VCC with a bottom-up approach. Our starting point is CompCertO, an extension of CompCert -- the state-of-the-art verified compiler -- that supports VCC but lacks a unified interface. We discover that a CompCert Kripke Logical Relation (CKLR) in CompCertO provides a uniform notion of memory protection for evolving memory states across modules and is transitively composable. Based on this uniform and composable CKLR, we then merge the simulation relations for all the compiler pass in CompCertO (except for three value analysis passes) into a unified interface. We demonstrate the conciseness and effectiveness of this unified interface by applying it to verify the compositional compilation of a non-trivial heterogeneous program with mutual recursion.
相關內容
Distributed Linearly Separable Computation problem under the cyclic assignment is studied in this paper. It is a problem widely existing in cooperated distributed gradient coding, real-time rendering, linear transformers, etc. In a distributed computing system, a master asks $N$ distributed workers to compute a linearly separable function from $K$ datasets. The task function can be expressed as $K_c$ linear combinations of $K$ messages, where each message is the output of one individual function of one dataset. Straggler effect is also considered, such that from the answers of each $N_r$ worker, the master should recover the task. The computation cost is defined as the number of datasets assigned to each worker, while the communication cost is defined as the number of (coded) messages which should be received. The objective is to characterize the optimal tradeoff between the computation and communication costs. Various distributed computing scheme were proposed in the literature with a well-known cyclic data assignment, but the (order) optimality of this problem remains open, even under the cyclic assignment. This paper proposes a new computing scheme with the cyclic assignment based on interference alignment, which is near optimal under the cyclic assignment.
Achieving high-quality semantic segmentation predictions using only image-level labels enables a new level of real-world applicability. Although state-of-the-art networks deliver reliable predictions, the amount of handcrafted pixel-wise annotations to enable these results are not feasible in many real-world applications. Hence, several works have already targeted this bottleneck, using classifier-based networks like Class Activation Maps~\cite{CAM} (CAMs) as a base. Addressing CAM's weaknesses of fuzzy borders and incomplete predictions, state-of-the-art approaches rely only on adding regulations to the classifier loss or using pixel-similarity-based refinement after the fact. We propose a framework that introduces an additional module using object perimeters for improved saliency. We define object perimeter information as the line separating the object and background. Our new PerimeterFit module will be applied to pre-refine the CAM predictions before using the pixel-similarity-based network. In this way, our PerimeterFit increases the quality of the CAM prediction while simultaneously improving the false negative rate. We investigated a wide range of state-of-the-art unsupervised semantic segmentation networks and edge detection techniques to create useful perimeter maps, which enable our framework to predict object locations with sharper perimeters. We achieved up to 1.5% improvement over frameworks without our PerimeterFit module. We conduct an exhaustive analysis to illustrate that SILOP enhances existing state-of-the-art frameworks for image-level-based semantic segmentation. The framework is open-source and accessible online at //github.com/ErikOstrowski/SILOP.
Self-supervised facial representation has recently attracted increasing attention due to its ability to perform face understanding without relying on large-scale annotated datasets heavily. However, analytically, current contrastive-based self-supervised learning (SSL) still performs unsatisfactorily for learning facial representation. More specifically, existing contrastive learning (CL) tends to learn pose-invariant features that cannot depict the pose details of faces, compromising the learning performance. To conquer the above limitation of CL, we propose a novel Pose-disentangled Contrastive Learning (PCL) method for general self-supervised facial representation. Our PCL first devises a pose-disentangled decoder (PDD) with a delicately designed orthogonalizing regulation, which disentangles the pose-related features from the face-aware features; therefore, pose-related and other pose-unrelated facial information could be performed in individual subnetworks and do not affect each other's training. Furthermore, we introduce a pose-related contrastive learning scheme that learns pose-related information based on data augmentation of the same image, which would deliver more effective face-aware representation for various downstream tasks. We conducted linear evaluation on four challenging downstream facial understanding tasks, ie, facial expression recognition, face recognition, AU detection and head pose estimation. Experimental results demonstrate that our method significantly outperforms state-of-the-art SSL methods. Code is available at //github.com/DreamMr/PCL}{//github.com/DreamMr/PCL
In recent years, there is an increasing interests in reconstruction based generative models for image One-Class Novelty Detection, most of which only focus on image-level information. While in this paper, we further exploit the latent space of Variational Auto-encoder (VAE), a typical reconstruction based model, and we innovatively divide it into three regions: Normal/Anomalous/Unknown-semantic-region. Based on this hypothesis, we propose a new VAE architecture, Recoding Semantic Consistency Based VAE (RSC-VAE), combining VAE with recoding mechanism and constraining the semantic consistency of two encodings. We come up with three training modes of RSC-VAE: 1. One-Class Training Mode, alleviating False Positive problem of normal samples; 2. Distributionally-Shifted Training Mode, alleviating False Negative problem of anomalous samples; 3. Extremely-Imbalanced Training Mode, introducing a small number of anomalous samples for training to enhance the second mode. The experimental results on multiple datasets demonstrate that our mechanism achieves state-of-the-art performance in various baselines including VAE.
The intricate interplay of source dynamics, unreliable channels, and staleness of information has long been recognized as a significant impediment for the receiver to achieve accurate, timely, and most importantly, goal-oriented decision making. Thus, a plethora of promising metrics, such as Age of Information, Value of Information, and Mean Square Error, have emerged to quantify these underlying adverse factors. Following this avenue, optimizing these metrics has indirectly improved the utility of goal-oriented decision making. Nevertheless, no metric has hitherto been expressly devised to evaluate the utility of a goal-oriented decision-making process. To this end, this paper investigates a novel performance metric, the Goal-oriented Tensor (GoT), to directly quantify the impact of semantic mismatches on the goal-oriented decision making. Based on the GoT, we consider a sampler-decision maker pair that work collaboratively and distributively to achieve a shared goal of communications. We formulate an infinite-horizon Decentralized Partially Observable Markov Decision Process (Dec-POMDP) to conjointly deduce the optimal deterministic sampling policy and decision-making policy. The simulation results reveal that the sampler-decision maker co-design surpasses beyond the current literature on AoI and its variants in terms of both goal achievement utility and sparse sampling rate, signifying a notable accomplishment for a sparse sampler and goal-oriented decision maker co-design.
Multi-stage ranking pipelines have been a practical solution in modern search systems, where the first-stage retrieval is to return a subset of candidate documents, and latter stages attempt to re-rank those candidates. Unlike re-ranking stages going through quick technique shifts during past decades, the first-stage retrieval has long been dominated by classical term-based models. Unfortunately, these models suffer from the vocabulary mismatch problem, which may block re-ranking stages from relevant documents at the very beginning. Therefore, it has been a long-term desire to build semantic models for the first-stage retrieval that can achieve high recall efficiently. Recently, we have witnessed an explosive growth of research interests on the first-stage semantic retrieval models. We believe it is the right time to survey current status, learn from existing methods, and gain some insights for future development. In this paper, we describe the current landscape of the first-stage retrieval models under a unified framework to clarify the connection between classical term-based retrieval methods, early semantic retrieval methods and neural semantic retrieval methods. Moreover, we identify some open challenges and envision some future directions, with the hope of inspiring more researches on these important yet less investigated topics.
Images can convey rich semantics and induce various emotions in viewers. Recently, with the rapid advancement of emotional intelligence and the explosive growth of visual data, extensive research efforts have been dedicated to affective image content analysis (AICA). In this survey, we will comprehensively review the development of AICA in the recent two decades, especially focusing on the state-of-the-art methods with respect to three main challenges -- the affective gap, perception subjectivity, and label noise and absence. We begin with an introduction to the key emotion representation models that have been widely employed in AICA and description of available datasets for performing evaluation with quantitative comparison of label noise and dataset bias. We then summarize and compare the representative approaches on (1) emotion feature extraction, including both handcrafted and deep features, (2) learning methods on dominant emotion recognition, personalized emotion prediction, emotion distribution learning, and learning from noisy data or few labels, and (3) AICA based applications. Finally, we discuss some challenges and promising research directions in the future, such as image content and context understanding, group emotion clustering, and viewer-image interaction.
Knowledge is a formal way of understanding the world, providing a human-level cognition and intelligence for the next-generation artificial intelligence (AI). One of the representations of knowledge is the structural relations between entities. An effective way to automatically acquire this important knowledge, called Relation Extraction (RE), a sub-task of information extraction, plays a vital role in Natural Language Processing (NLP). Its purpose is to identify semantic relations between entities from natural language text. To date, there are several studies for RE in previous works, which have documented these techniques based on Deep Neural Networks (DNNs) become a prevailing technique in this research. Especially, the supervised and distant supervision methods based on DNNs are the most popular and reliable solutions for RE. This article 1)introduces some general concepts, and further 2)gives a comprehensive overview of DNNs in RE from two points of view: supervised RE, which attempts to improve the standard RE systems, and distant supervision RE, which adopts DNNs to design the sentence encoder and the de-noise method. We further 3)cover some novel methods and describe some recent trends and discuss possible future research directions for this task.
External knowledge is often useful for natural language understanding tasks. We introduce a contextual text representation model called Conceptual-Contextual (CC) embeddings, which incorporates structured knowledge into text representations. Unlike entity embedding methods, our approach encodes a knowledge graph into a context model. CC embeddings can be easily reused for a wide range of tasks just like pre-trained language models. Our model effectively encodes the huge UMLS database by leveraging semantic generalizability. Experiments on electronic health records (EHRs) and medical text processing benchmarks showed our model gives a major boost to the performance of supervised medical NLP tasks.
Semantic Role Labeling (SRL) is believed to be a crucial step towards natural language understanding and has been widely studied. Recent years, end-to-end SRL with recurrent neural networks (RNN) has gained increasing attention. However, it remains a major challenge for RNNs to handle structural information and long range dependencies. In this paper, we present a simple and effective architecture for SRL which aims to address these problems. Our model is based on self-attention which can directly capture the relationships between two tokens regardless of their distance. Our single model achieves F$_1=83.4$ on the CoNLL-2005 shared task dataset and F$_1=82.7$ on the CoNLL-2012 shared task dataset, which outperforms the previous state-of-the-art results by $1.8$ and $1.0$ F$_1$ score respectively. Besides, our model is computationally efficient, and the parsing speed is 50K tokens per second on a single Titan X GPU.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191