搜索结果：cheap

As Transfer Learning from large-scale pre-trained models becomes more prevalent in Natural Language Processing (NLP), operating these large models in on-the-edge and/or under constrained computational training or inference budgets remains challenging. In this work, we propose a method to pre-train a smaller general-purpose language representation model, called DistilBERT, which can then be fine-tuned with good performances on a wide range of tasks like its larger counterparts. While most prior work investigated the use of distillation for building task-specific models, we leverage knowledge distillation during the pre-training phase and show that it is possible to reduce the size of a BERT model by 40%, while retaining 97% of its language understanding capabilities and being 60% faster. To leverage the inductive biases learned by larger models during pre-training, we introduce a triple loss combining language modeling, distillation and cosine-distance losses. Our smaller, faster and lighter model is cheaper to pre-train and we demonstrate its capabilities for on-device computations in a proof-of-concept experiment and a comparative on-device study.

A significant fraction of software failures in large-scale Internet systems are cured by rebooting, even when the exact failure causes are unknown. However, rebooting can be expensive, causing nontrivial service disruption or downtime even when clusters and failover are employed. In this work we separate process recovery from data recovery to enable microrebooting -- a fine-grain technique for surgically recovering faulty application components, without disturbing the rest of the application. We evaluate microrebooting in an Internet auction system running on an application server. Microreboots recover most of the same failures as full reboots, but do so an order of magnitude faster and result in an order of magnitude savings in lost work. This cheap form of recovery engenders a new approach to high availability: microreboots can be employed at the slightest hint of failure, prior to node failover in multi-node clusters, even when mistakes in failure detection are likely; failure and recovery can be masked from end users through transparent call-level retries; and systems can be rejuvenated by parts, without ever being shut down.

A large body of literature suggests willingness‐to‐pay is overstated in hypothetical valuation questions as compared to when actual payment is required. Recently, “cheap talk” has been proposed to eliminate the potential bias in hypothetical valuation questions. Cheap talk refers to process of explaining hypothetical bias to individuals prior to asking a valuation question. This study explores the effect of cheap talk in a mass mail survey using a conventional value elicitation technique. Results indicate that cheap talk was effective at reducing willingness‐to‐pay for most survey participants; however, consistent with previous research, cheap talk did not reduce willingness‐to‐pay for knowledgeable consumers.

Nature, money, work, care, food, energy, and lives: these are the seven things that have made our world and will shape its future. In making these things cheap, modern commerce has transformed, governed, and devastated Earth. In A History of the World in Seven Cheap Things , Raj Patel and Jason W. Moore present a new approach to analyzing today’s planetary emergencies. Bringing the latest ecological research together with histories of colonialism, indigenous struggles, slave revolts, and other rebellions and uprisings, Patel and Moore demonstrate that throughout history, crises have always prompted fresh strategies to make the world cheap and safe for capitalism. At a time of crisis in all seven cheap things, innovative and systemic thinking is urgently required. This book proposes a radical new way of understanding—and reclaiming—the planet in the turbulent twenty-first century.

Economists often ask how private information is shared through markets, costly signaling, and other mechanisms. Yet most information sharing is done through ordinary, informal talk. Economists are inconsistent in their view of such ‘cheap talk’: sometimes it is supposed that communication generally leads to efficient equilibria; other times it is supposed that since ‘talk is cheap,’ it is never credible. The authors think both views are wrong. In this paper, they describe what some recent research in game theory teaches about when people will convey private information by cheap talk.

Deploying convolutional neural networks (CNNs) on embedded devices is difficult due to the limited memory and computation resources. The redundancy in feature maps is an important characteristic of those successful CNNs, but has rarely been investigated in neural architecture design. This paper proposes a novel Ghost module to generate more feature maps from cheap operations. Based on a set of intrinsic feature maps, we apply a series of linear transformations with cheap cost to generate many ghost feature maps that could fully reveal information underlying intrinsic features. The proposed Ghost module can be taken as a plug-and-play component to upgrade existing convolutional neural networks. Ghost bottlenecks are designed to stack Ghost modules, and then the lightweight GhostNet can be easily established. Experiments conducted on benchmarks demonstrate that the proposed Ghost module is an impressive alternative of convolution layers in baseline models, and our GhostNet can achieve higher recognition performance (e.g. 75.7% top-1 accuracy) than MobileNetV3 with similar computational cost on the ImageNet ILSVRC-2012 classification dataset. Code is available at https://github.com/huawei-noah/ghostnet.

With cheap talk, more can be achieved by long conversations than by a single message—even when one side is strictly better informed than the other. (“Cheap talk” means plain conversation—unmediated, nonbinding, and payoff-irrelevant.) This work characterizes the equilibrium payoffs for all two-person games in which one side is better informed than the other and cheap talk is permitted.

Conventionally, Apartheid is regarded as no more than an intensification of the earlier policy of Segregation and is ascribed simplistically to the particular racial ideology of the ruling Nationalist Party. In this article substantial differences between Apartheid and Segregation are identified and explained by reference to the changing relations of capitalist and African pre-capitalist modes of production. The supply of African migrant labour-power, at a wage below its cost of reproduction, is a function of the existence of the pre-capitalist mode. The dominant capitalist mode of production tends to dissolve the pre-capitalist mode thus threatening the conditions of reproduction of cheap migrant labour-power and thereby generating intense conflict against the system of Segregation. In these conditions Segregation gives way to Apartheid which provides the specific mechanism for maintaining labour-power cheap through the elaboration of the entire system of domination and control and the transformation of the function of the pre-capitalist societies.

We show how costless, nonbinding, nonverifiable communication (cheap talk) can achieve partial coordination among potential entrants into a natural-monopoly industry, where the payoffs are qualitatively like the "battle of the sexes." The analysis would apply equally in other economic situations with such payoffs, for example, bargaining under complete information or choosing compatibility standards. While cheap talk helps achieve asymmetric coordination in a symmetric mixed-strategy equilibrium, it cannot achieve complete coordination if the game involves even a small amount of conflict.

In recent decades, cheap labor has played a central role in the Chinese model, which has relied on expanded participation in world trade as a main driver of growth. At the beginning of China's economic reforms in 1978, the annual wage of a Chinese urban worker was only $1,004 in U.S. dollars. The Chinese wage was only 3 percent of the average U.S. wage at that time, and it was also significantly lower than the wages in neighboring Asian countries such as the Philippines and Thailand. The Chinese wage was also low relative to productivity. However, wages are now rising in China. In 2010, the annual wage of a Chinese urban worker reached $5,487 in U.S. dollars, which is similar to wages earned by workers in the Philippines and Thailand and significantly higher than those earned by workers in India and Indonesia. China's wages also increased faster than productivity since the late 1990s, suggesting that Chinese labor is becoming more expensive in this sense as well. The increase in China's wages is not confined to any sector, as wages have increased for both skilled and unskilled workers, for both coastal and inland areas, and for both exporting and nonexporting firms. We benchmark wage growth to productivity growth using both national- and industry-level data, showing that Chinese labor was kept cheap until the late 1990s but the relative cost of labor has increased since then. Finally, we discuss the main forces that are pushing wages up.

Human linguistic annotation is crucial for many natural language processing tasks but can be expensive and time-consuming. We explore the use of Amazon's Mechanical Turk system, a significantly cheaper and faster method for collecting annotations from a broad base of paid non-expert contributors over the Web. We investigate five tasks: affect recognition, word similarity, recognizing textual entailment, event temporal ordering, and word sense disambiguation. For all five, we show high agreement between Mechanical Turk non-expert annotations and existing gold standard labels provided by expert labelers. For the task of affect recognition, we also show that using non-expert labels for training machine learning algorithms can be as effective as using gold standard annotations from experts. We propose a technique for bias correction that significantly improves annotation quality on two tasks. We conclude that many large labeling tasks can be effectively designed and carried out in this method at a fraction of the usual expense.

Unbiased Value Estimates for Environmental Goods: A Cheap Talk Design for the Contingent Valuation Method by Ronald G. Cummings and Laura O. Taylor. Published in volume 89, issue 3, pages 649-665 of American Economic Review, June 1999

There are typically multiple equilibrium outcomes in the Crawford–Sobel (CS) model of strategic information transmission. This paper identifies a simple condition on equilibrium payoffs, called NITS (no incentive to separate), that selects among CS equilibria. Under a commonly used regularity condition, only the equilibrium with the maximal number of induced actions satisfies NITS. We discuss various justifications for NITS, including perturbed cheap-talk games with nonstrategic players or costly lying. We also apply NITS to other models of cheap talk, illustrating its potential beyond the CS framework.

We consider the credibility, persuasiveness, and informativeness of multidimensional cheap talk by an expert to a decision maker. We find that an expert with state-independent preferences can always make credible comparative statements that trade off the expert's incentive to exaggerate on each dimension. Such communication benefits the expert—cheap talk is “persuasive”—if her preferences are quasiconvex. Communication benefits a decision maker by allowing for a more informed decision, but strategic interactions between multiple decision makers can reverse this gain. We apply these results to topics including product recommendations, voting, auction disclosure, and advertising. (JEL D44, D72, D82, D83, M37)

Contents Acknowledgments Introduction 1. The Homosocial World of Working-Class Amusements 2. Leisure and Labor 3. Putting on Style 4. Dance Madness 5. The Coney Island Excursion 6. Cheap Theater and the Nickel Dumps 7. Reforming Working Women's Recreation Conclusion Notes Index

We present Rsubread, a Bioconductor software package that provides high-performance alignment and read counting functions for RNA-seq reads. Rsubread is based on the successful Subread suite with the added ease-of-use of the R programming environment, creating a matrix of read counts directly as an R object ready for downstream analysis. It integrates read mapping and quantification in a single package and has no software dependencies other than R itself. We demonstrate Rsubread's ability to detect exon-exon junctions de novo and to quantify expression at the level of either genes, exons or exon junctions. The resulting read counts can be input directly into a wide range of downstream statistical analyses using other Bioconductor packages. Using SEQC data and simulations, we compare Rsubread to TopHat2, STAR and HTSeq as well as to counting functions in the Bioconductor infrastructure packages. We consider the performance of these tools on the combined quantification task starting from raw sequence reads through to summary counts, and in particular evaluate the performance of different combinations of alignment and counting algorithms. We show that Rsubread is faster and uses less memory than competitor tools and produces read count summaries that more accurately correlate with true values.

Abstract Although globalization has usually been associated with advanced communications technology, arguably nothing has facilitated global linkage more than the boom in ordinary, cheap international telephone calls. Low‐cost calls serve as a kind of social glue connecting small‐scale social formations across the globe. In this article I present recent data on the rapid growth and diffusion of telephone traffic and describe the proliferation of prepaid phonecards. Second, I outline the commercial, social and geographical ramifications of this explosion in transnational communication.

We consider the problems of societal norms for cooperation and reputation when it is possible to obtain cheap pseudonyms, something that is becoming quite common in a wide variety of interactions on the Internet. This introduces opportunities to misbehave without paying reputational consequences. A large degree of cooperation can still emerge, through a convention in which newcomers “pay their dues” by accepting poor treatment from players who have established positive reputations. One might hope for an open society where newcomers are treated well, but there is an inherent social cost in making the spread of reputations optional. We prove that no equilibrium can sustain significantly more cooperation than the dues‐paying equilibrium in a repeated random matching game with a large number of players in which players have finite lives and the ability to change their identities, and there is a small but nonvanishing probability of mistakes. Although one could remove the inefficiency of mistreating newcomers by disallowing anonymity, this is not practical or desirable in a wide variety of transactions. We discuss the use of entry fees, which permits newcomers to be trusted but excludes some players with low payoffs, thus introducing a different inefficiency. We also discuss the use of free but unreplaceable pseudonyms, and describe a mechanism that implements them using standard encryption techniques, which could be practically implemented in electronic transactions.

It has become common to distribute software in forms that are isomorphic to the original source code. An important example is Java bytecode. Since such codes are easy to decompile, they increase the risk of malicious reverse engineering attacks.In this paper we describe the design of a Java code obfuscator, a tool which - through the application of code transformations - converts a Java program into an equivalent one that is more difficult to reverse engineer.We describe a number of transformations which obfuscate control-flow. Transformations are evaluated with respect to potency (To what degree is a human reader confused?), resilience (How well are automatic deobfuscation attacks resisted?), cost (How much time/space overhead is added?), and stealth (How well does obfuscated code blend in with the original code?).The resilience of many control-altering transformations rely on the resilience of opaque predicates. These are boolean valued expressions whose values are known to the obfuscator but difficult to determine for an automatic deobfuscator. We show how to construct resilient, cheap, and stealthy opaque predicates based on the intractability of certain static analysis problems such as alias analysis.