Equivalent Comparisons of Experiments
By D. Blackwell. 1953
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
- Peter Kairouz
- Sewoong Oh
- Zinan Lin
- Alexander Shapiro
- Pramod Viswanath
- Ashish Khetan
- Andrzej Ruszczynski
- Francis R. Bach
Citations in the corpus
- Extremal Mechanisms for Local Differential Privacy (2014, positive): “This is established by proving a very strong statistical dominance using Blackwell’s celebrated result on comparisons of statistical experiments [4].”
- The Composition Theorem for Differential Privacy (2013, background): “…and ˜ δ ∈ [0 , 1] , the class of ( ε (cid:96) , δ (cid:96) ) - differentially private mechanisms satisfy ˜ ε ˜ δ , 1 This tells us that the ε (cid:96) ’s sum up under composition : whenever we have kε or kε 2 in (7) we can replace it by the summation to get the general result for heterogeneous case.”
- PacGAN: The Power of Two Samples in Generative Adversarial Networks (2017, background): “This allows us to bring powerful mathematical tools from this mature area in statistics and information theory—in particular, the data processing inequalities originating from the seminal work of Blackwell [6].”
- The Politics of Attention (2018, unknown): “We adopt Blackwell (1953)’s notion of informativeness: Definition 5. f is more Blackwell-informative than f ′ ( f ′ is a garble of f , f f ′) if there exists a Markov kenel ρ such that for all a and ω′, f ′ (ω′ | a) = ∑ ω∈Ω f (ω | a) ρ (ω′ | ω) .”
- Quantifying unique information (2013, background): “This result is known as Blackwell’s theorem [9]; see also [8].”
- Reputation Mechanism Design in Online Trading Environments with Pure Moral Hazard (2005, background): “Common wisdom in decision theory suggests that more information leads to better outcomes (Blackwell 1951, 1953).”
- Information Theory and an Extension of the Maximum Likelihood Principle (1973, background): “It can be shown [9] that for the purpose of discrimination between the two probability distributions with density functions h(x) (i = 0, 1) all the necessary information are contained in the likelihood ratio T(x) = 11 (x)/Io(x) in the sense that any decision procedure with a prescribed loss of discr”
- Persuasion of a Privately Informed Receiver (2017, unknown): “Both Blackwell (1953) and Gutmann et al. (1991) are based on the majorization theory initiated by Hardy et al. (1929).”
- Optimization of Convex Risk Functions (2006, background): “If X 1 icx X 2 , by Müller and Stoyan [15, Corollary 1.5.21] (see also Blackwell [3], Strassen [29]), we can find This means that is consistent with the increasing convex order.”
- Suspense and Surprise (2015, unknown): “This definition is motivated by Blackwell ð1953Þ: assuming that W is concave is equivalent to assuming that receiving information must, on average, reduce uncertainty.”
- On the Simple Economics of Advertising, Marketing, and Product Design (2006, background): “50Similarly, there is no general result relating Blackwell (1953) ordered advertising-signal distributions to quasi-convexity of payoffs.”
- Binary Payment Schemes: Moral Hazard and Loss Aversion (2010, unknown): “Their estimates suggest that a reference-dependent model of drivers’ labor supply where targets are carefully 5The feature that the reference point is determined by the decision maker’s forward-looking expectations is shared with the disappointment aversion models of David E. Bell (1985), Graham Loo”
- The Impact of Supervisory Board Composition on CSR Reporting. Evidence from the German Two-Tier System (2016, background): “Additional (sustainability) information has the potential to be routinely beneficial [13] in that a higher intensity in sustainability reporting might correlate with more informed stakeholder decisions and their ability to influence the company value in a positive way.”