On the optimal boolean function for prediction under quadratic loss

Research output: Contribution to journalArticlepeer-review


Suppose Yn is obtained by observing a uniform Bernoulli random vector Xn through a binary symmetric channel. Courtade and Kumar asked how large the mutual information between Yn and a Boolean function b(Xn) could be, and conjectured that the maximum is attained by a dictator function. An equivalent formulation of this conjecture is that dictator minimizes the prediction cost in a sequential prediction of Yn under logarithmic loss, given b(Xn). In this paper, we study the question of minimizing the sequential prediction cost under a different (proper) loss function the quadratic loss. In the noiseless case, we show that majority asymptotically minimizes this prediction cost among all Boolean functions.We further show that for weak noise, majority is better than a dictator, and that for a strong noise dictator outperforms majority. We conjecture that for quadratic loss, there is no single sequence of Boolean functions that is simultaneously (asymptotically) optimal at all noise levels.

Original languageEnglish
Article number7886288
Pages (from-to)4202-4217
Number of pages16
JournalIEEE Transactions on Information Theory
Issue number7
StatePublished - Jul 2017


  • Boolean functions
  • Logarithmic loss function
  • Pinsker's inequality
  • Quadratic loss function
  • Sequential prediction

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences


Dive into the research topics of 'On the optimal boolean function for prediction under quadratic loss'. Together they form a unique fingerprint.

Cite this