Corrigendum: Experimental Certification of Random Numbers via Quantum Contextuality

This corrects the article DOI: 10.1038/srep01627.

The original version of this Article contained an error in the spelling of the author Yangchao Shen, which was incorrectly given as Shen Yangchao.
This error has now been corrected in the PDF and HTML versions of the Article.
In addition, the 〈 V i V j 〉 terms in Table 1 were omitted from the calculation of χ′ KCBS in Equation 4. Therefore, in Table 1 "In our experiment, we generate 1 × 10 5 random numbers that are guaranteed to have 5.2 × 10 4 bits of minimum entropy with a 99% confidence level. " should read: "In our experiment, we generate 1 × 10 5 random numbers that are guaranteed to have 2.4 × 10 4 bits of minimum entropy with a 99% confidence level. " In the Results section, under subheading 'Random number results' , "As shown in Table 1 In the title of Table 1, "Our experimental test clearly shows the violation of the extended inequality (3) with 31 σ " should read: "Our experimental test clearly shows the violation of the extended inequality (3) with 18 σ " Moreover, the presented data for the biased choice of measurement settings does not show the net randomness after including the terms 〈 V i V j 〉 in Table 1 for the χ′ KCBS in Equation 4. it is necessary to double the total experimental round as n = 2 × 10 5 with the new biased distribution parameter α = 12 in order to observe the net randomness. Therefore, the contents of the paper related to the biased choice of measurement settings should be corrected as follows.
In the Results section, under subheading 'Random number results' , "We also generate random bits with a biased choice of measurement settings, where P (V 1 ) = 1 − 4q, P (V 2 ) = P (V 3 ) = P (V 4 ) = P (V 5 ) = q, and q = αn −1/2 with α = 6 and n = 10 5 . We observe basically the same behavior of the min-entropy for the generated stream except for a slightly smaller bound due to the non-uniform setting. We get the min-entropy bound > . × ∞ H v V ( ) 1 4 10 bia 4 from 1 × 10 5 rounds with violation of ˆ= .
L 3 692. For the biased choice of measurement settings, the output entropy (3.95 × 10 4 ) exceeds the input entropy (3.28 × 10 4 ), and we obtain 6.8 × 10 3 net random bits. " In the legend of   In addition, this Article contains typographical errors in the Results section, under subheading 'The KCBS inequality' .