Timezone: »

Variational Inference and Model Selection with Generalized Evidence Bounds
Liqun Chen · Chenyang Tao · RUIYI (ROY) ZHANG · Ricardo Henao · Lawrence Carin

Wed Jul 11 05:30 AM -- 05:50 AM (PDT) @ A4

Recent advances on the scalability and flexibility of variational inference have made it successful at unravelling hidden patterns in complex data. In this work we propose a new variational bound formulation, yielding an estimator that extends beyond the conventional variational bound. It naturally subsumes the importance-weighted and Renyi bounds as special cases, and it is provably sharper than these counterparts. We also present an improved estimator for variational learning, and advocate a novel high signal-to-variance ratio update rule for the variational parameters. We discuss model-selection issues associated with existing evidence-lower-bound-based variational inference procedures, and show how to leverage the flexibility of our new formulation to address them. Empirical evidence is provided to validate our claims.

Author Information

Liqun Chen (Duke University)
Chenyang Tao (Duke University)
RUIYI (ROY) ZHANG (Duke University)
Ricardo Henao (Duke University)
Lawrence Carin (Duke)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors