1,015 Hits in 2.0 sec

Informational Substitutes [article]

Yiling Chen, Bo Waggoner
2017 arXiv   pre-print
Chen et al. [2007] and Dimitrov and Sami [2008] were combined and extended in Chen et al. [2010] .  ...  This is the approach of submodular surrogates in the literature [Chen et al., 2015] .  ... 
arXiv:1703.08636v1 fatcat:77cpin3pyrgipanu7l55f255xy

Privacy Games [article]

Yiling Chen and Or Sheffet and Salil Vadhan
2014 arXiv   pre-print
Chen et al [CCK + 13] proposed a refinement where the privacy loss is measured with respect to the given input and output.  ... 
arXiv:1410.1920v1 fatcat:nkdk3j3u2rg7hfalwoifqza3ne

Optimal Scoring Rule Design [article]

Yiling Chen, Fang-Yi Yu
2021 arXiv   pre-print
This paper introduces an optimization problem for proper scoring rule design. Consider a principal who wants to collect an agent's prediction about an unknown state. The agent can either report his prior prediction or access a costly signal and report the posterior prediction. Given a collection of possible distributions containing the agent's posterior prediction distribution, the principal's objective is to design a bounded scoring rule to maximize the agent's worst-case payoff increment
more » ... en reporting his posterior prediction and reporting his prior prediction. We study two settings of such optimization for proper scoring rules: static and asymptotic settings. In the static setting, where the agent can access one signal, we propose an efficient algorithm to compute an optimal scoring rule when the collection of distributions is finite. The agent can adaptively and indefinitely refine his prediction in the asymptotic setting. We first consider a sequence of collections of posterior distributions with vanishing covariance, which emulates general estimators with large samples, and show the optimality of the quadratic scoring rule. Then, when the agent's posterior distribution is a Beta-Bernoulli process, we find that the log scoring rule is optimal. We also prove the optimality of the log scoring rule over a smaller set of functions for categorical distributions with Dirichlet priors.
arXiv:2107.07420v1 fatcat:myaymjgtibhkffj64baqzkl5iq

Elicitation for Aggregation [article]

Rafael M. Frongillo, Yiling Chen, Ian A. Kash
2014 arXiv   pre-print
However, the dynamic nature of these mechanisms can induce complicated strategic play and obfuscate individual-level information (Hansen, Schmidt, and Strobel 2001; Chen et al. 2010; Gao, Zhang, and Chen  ...  To reduce the total payment of the principal, researchers design shared scoring rules (Kilgour and Gerchak 2004; Johnstone 2007) and wagering mechanisms (Lambert et al. 2008; Lambert et al. 2014; Chen  ... 
arXiv:1410.0375v1 fatcat:m3a2wve4pjcojckvbr7e64jpwm

Fair Classification and Social Welfare [article]

Lily Hu, Yiling Chen
2019 arXiv   pre-print
Now that machine learning algorithms lie at the center of many resource allocation pipelines, computer scientists have been unwittingly cast as partial social planners. Given this state of affairs, important questions follow. What is the relationship between fairness as defined by computer scientists and notions of social welfare? In this paper, we present a welfare-based analysis of classification and fairness regimes. We translate a loss minimization program into a social welfare maximization
more » ... problem with a set of implied welfare weights on individuals and groups--weights that can be analyzed from a distribution justice lens. In the converse direction, we ask what the space of possible labelings is for a given dataset and hypothesis class. We provide an algorithm that answers this question with respect to linear hyperplanes in R^d that runs in O(n^dd). Our main findings on the relationship between fairness criteria and welfare center on sensitivity analyses of fairness-constrained empirical risk minimization programs. We characterize the ranges of Δϵ perturbations to a fairness parameter ϵ that yield better, worse, and neutral outcomes in utility for individuals and by extension, groups. We show that applying more strict fairness criteria that are codified as parity constraints, can worsen welfare outcomes for both groups. More generally, always preferring "more fair" classifiers does not abide by the Pareto Principle---a fundamental axiom of social choice theory and welfare economics. Recent work in machine learning has rallied around these notions of fairness as critical to ensuring that algorithmic systems do not have disparate negative impact on disadvantaged social groups. By showing that these constraints often fail to translate into improved outcomes for these groups, we cast doubt on their effectiveness as a means to ensure justice.
arXiv:1905.00147v1 fatcat:2mcemfwdk5d3xapb3dndeui6ha

Welfare and Distributional Impacts of Fair Classification [article]

Lily Hu, Yiling Chen
2018 arXiv   pre-print
Current methodologies in machine learning analyze the effects of various statistical parity notions of fairness primarily in light of their impacts on predictive accuracy and vendor utility loss. In this paper, we propose a new framework for interpreting the effects of fairness criteria by converting the constrained loss minimization problem into a social welfare maximization problem. This translation moves a classifier and its output into utility space where individuals, groups, and society
more » ... large experience different welfare changes due to classification assignments. Under this characterization, predictions and fairness constraints are seen as shaping societal welfare and distribution and revealing individuals' implied welfare weights in society--weights that may then be interpreted through a fairness lens. The social welfare formulation of the fairness problem brings to the fore concerns of distributive justice that have always had a central albeit more implicit role in standard algorithmic fairness approaches.
arXiv:1807.01134v1 fatcat:ymjjoza2o5gnba5mplexk54qcu

Fairness at Equilibrium in the Labor Market [article]

Lily Hu, Yiling Chen
2017 arXiv   pre-print
Recent literature on computational notions of fairness has been broadly divided into two distinct camps, supporting interventions that address either individual-based or group-based fairness. Rather than privilege a single definition, we seek to resolve both within the particular domain of employment discrimination. To this end, we construct a dual labor market model composed of a Temporary Labor Market, in which firm strategies are constrained to ensure group-level fairness, and a Permanent
more » ... or Market, in which individual worker fairness is guaranteed. We show that such restrictions on hiring practices induces an equilibrium that Pareto-dominates those arising from strategies that employ statistical discrimination or a "group-blind" criterion. Individual worker reputations produce externalities for collective reputation, generating a feedback loop termed a "self-fulfilling prophecy." Our model produces its own feedback loop, raising the collective reputation of an initially disadvantaged group via a fairness intervention that need not be permanent. Moreover, we show that, contrary to popular assumption, the asymmetric equilibria resulting from hiring practices that disregard group-fairness may be immovable without targeted intervention. The enduring nature of such equilibria that are both inequitable and Pareto inefficient suggest that fairness interventions are of critical importance in moving the labor market to be more socially just and efficient.
arXiv:1707.01590v1 fatcat:3dtjfyhgtbgillrymrl5e6wagq

Designing Informative Securities [article]

Yiling Chen, Mike Ruberry, Jennifer Wortman Vaughan
2012 arXiv   pre-print
We create a formal framework for the design of informative securities in prediction markets. These securities allow a market organizer to infer the likelihood of events of interest as well as if he knew all of the traders' private signals. We consider the design of markets that are always informative, markets that are informative for a particular signal structure of the participants, and informative markets constructed from a restricted selection of securities. We find that to achieve
more » ... eness, it can be necessary to allow participants to express information that may not be directly of interest to the market organizer, and that understanding the participants' signal structure is important for designing informative prediction markets.
arXiv:1210.4837v1 fatcat:kpmgmd5spba6ll4msezajnepee

Learning to Incentivize: Eliciting Effort via Output Agreement [article]

Yang Liu, Yiling Chen
2016 arXiv   pre-print
Ming Yin and Yiling Chen. Bonus or Not? Learn to Reward in Crowdsourcing.  ...  This problem is called information elicitation without verification (IEWV) Waggoner and Chen [2014] .  ... 
arXiv:1604.04928v1 fatcat:aqydckycfvclbab7z3mwho2qfi

Privacy Games [chapter]

Yiling Chen, Or Sheffet, Salil Vadhan
2014 Lecture Notes in Computer Science  
Chen et al [4] proposed a refinement where the privacy loss is measured with respect to the given input and output.  ... 
doi:10.1007/978-3-319-13129-0_30 fatcat:7ullwuopk5bjbji747b7ymjozq

Strategyproof Linear Regression in High Dimensions [article]

Yiling Chen, Chara Podimata, Ariel D. Procaccia, Nisarg Shah
2018 arXiv   pre-print
This paper is part of an emerging line of work at the intersection of machine learning and mechanism design, which aims to avoid noise in training data by correctly aligning the incentives of data sources. Specifically, we focus on the ubiquitous problem of linear regression, where strategyproof mechanisms have previously been identified in two dimensions. In our setting, agents have single-peaked preferences and can manipulate only their response variables. Our main contribution is the
more » ... y of a family of group strategyproof linear regression mechanisms in any number of dimensions, which we call generalized resistant hyperplane mechanisms. The game-theoretic properties of these mechanisms -- and, in fact, their very existence -- are established through a connection to a discrete version of the Ham Sandwich Theorem.
arXiv:1805.10693v1 fatcat:se47bl6pzff3vi3kl67zwaf3ga

Complexity of Combinatorial Market Makers [article]

Yiling Chen, Lance Fortnow, Nicolas Lambert, David M. Pennock, Jennifer Wortman
2008 arXiv   pre-print
Subset Betting As in Chen et al.  ...  Chen et al. [4] analyze the the auctioneer matching problem for betting on permutations, examining two bidding languages.  ...  Our results on subset betting in LMSR contrast with those of Chen et al.  ... 
arXiv:0802.1362v1 fatcat:taxeaucm2jdwnenfsjdsjyckxy

Genomic analysis for heavy metal resistance inS. maltophilia [article]

Wenbang Yu, Xiaoxiao Chen, Yiling Sheng
2018 bioRxiv   pre-print
Pho and S. maltophilia K279a were conducted using the same genome annotation method as described above (Chen et al, 2018) .  ...  Then gene prediction, genome annotation and genome island identification were conducted using the method as described in the reference (Chen et al, 2018) .  ... 
doi:10.1101/404954 fatcat:mpmt4le7vbdaplrt7mo2erxlwy

An Optimization-Based Framework for Automated Market-Making [article]

Jacob Abernethy, Yiling Chen, Jennifer Wortman Vaughan
2010 arXiv   pre-print
For complete markets, Chen and Pennock [7] use the inverse of ∂ 2 C(q)/∂q 2 o to capture this notion for each security o independently.  ...  Similar conditions were suggested for complete markets by Chen and Vaughan [8] , who defined the notion of a valid cost function, and by Othman et al.  ... 
arXiv:1011.1941v1 fatcat:jkgkpilcsjey7f2lxq4442xor4

Cooperation in Threshold Public Projects with Binary Actions [article]

Yiling Chen, Biaoshuai Tao, Fang-Yi Yu
2021 arXiv   pre-print
When can cooperation arise from self-interested decisions in public goods games? And how can we help agents to act cooperatively? We examine these classical questions in a pivotal participation game, a variant of public good games, where heterogeneous agents make binary participation decisions on contributing their endowments, and the public project succeeds when it has enough contributions. We prove it is NP-complete to decide the existence of a cooperative Nash equilibrium such that the
more » ... t succeeds. We also identify two natural special scenarios where this decision problem is tractable. We then propose two algorithms to help cooperation in the game. Our first algorithm adds an external investment to the public project, and our second algorithm uses matching funds. We show that the cost to induce a cooperative Nash equilibrium is near-optimal for both algorithms. Finally, the cost of matching funds can always be smaller than the cost of adding an external investment. Intuitively, matching funds provide a greater incentive for cooperation than adding an external investment does.
arXiv:2105.08572v1 fatcat:d6mo5bmhevbjzcpa4adpnyyrxa
« Previous Showing results 1 — 15 out of 1,015 results