… Throughout the paper, when referring to a monotonic classifier, this signifies a fully monotonic
classifier. In addition, the interaction with a classifier is restricted to computing the value of …

… To achieve the goal, we utilize techniques for learning monotone Boolean functions. While
there … the problem of interactive learning of a monotone Boolean function from the complexity …

… of inferring the interaction network and the underlying functions simultaneously, accurately,
and … From application of our automated Boolean function learning method (results shown in …

… Boolean functions … monotonic Boolean functions, the Dedekind number [14], currently
known up to d = 8 [15]1. Whenever the specified k is lower than the maximum, Boolean functions …

… efficiently, we encode each Boolean function as a tree in … f2, where f1 and f2 are both monotone
Boolean formulae (ie, with … the interaction directions implied in the Boolean functions and …

… representation, reasoning, and interactive learning, and exploits the complementary strengths
of deep learning, non-monotonic logical reasoning with commonsense knowledge, and …

… Second, for the broad class of Fourier-sparse boolean functions, we show a multi-round (IP-…
class Td, in which each function is a concatenation of d monotone increasing thresholds. A …

… interaction between mutations in model genes (gain of Notch … monotonic whenever each
of its local functions is monotonic (… To test the impact of alternative Boolean functions, we …

… learning to measure the performance of a learning algorithm. In the classical single-agent
setting, online learning considers the following repeated interaction … selects a loss function ft(·), …

… as the deviation from a continuous additive regression function, then their … interaction via
the noncontinuous Boolean functions in the LSS model, and we derive consistent interaction …