Timezone: »

 
Poster
Probabilistic Rule Realization and Selection
Haizi Yu · Tianxi Li · Lav Varshney

Mon Dec 04 06:30 PM -- 10:30 PM (PST) @ Pacific Ballroom #21
in Posters Mon »

Abstraction and realization are bilateral processes that are key in deriving intelligence and creativity. In many domains, the two processes are approached through \emph{rules}: high-level principles that reveal invariances within similar yet diverse examples. Under a probabilistic setting for discrete input spaces, we focus on the rule realization problem which generates input sample distributions that follow the given rules. More ambitiously, we go beyond a mechanical realization that takes whatever is given, but instead ask for proactively selecting reasonable rules to realize. This goal is demanding in practice, since the initial rule set may not always be consistent and thus intelligent compromises are needed. We formulate both rule realization and selection as two strongly connected components within a single and symmetric bi-convex problem, and derive an efficient algorithm that works at large scale. Taking music compositional rules as the main example throughout the paper, we demonstrate our model's efficiency in not only music realization (composition) but also music interpretation and understanding (analysis).

Keywords: Model Selection and Structure Learning · Unsupervised Learning · Optimization · Representation Learning · Music Modeling and Analysis
[ Paper] 

Author Information

Haizi Yu (University of Illinois at Urbana-Champaign)
Tianxi Li (University of Michigan)
Lav Varshney (University of Illinois at Urbana-Champaign)

More from the Same Authors