Skip to main content

Bridging maximum likelihood and adversarial learning via α-divergence

Publication ,  Conference
Zhao, M; Cong, Y; Dai, S; Carin, L
Published in: AAAI 2020 - 34th AAAI Conference on Artificial Intelligence
January 1, 2020

Maximum likelihood (ML) and adversarial learning are two popular approaches for training generative models, and from many perspectives these techniques are complementary. ML learning encourages the capture of all data modes, and it is typically characterized by stable training. However, ML learning tends to distribute probability mass diffusely over the data space, e.g., yielding blurry synthetic images. Adversarial learning is well known to synthesize highly realistic natural images, despite practical challenges like mode dropping and delicate training. We propose an α-Bridge to unify the advantages of ML and adversarial learning, enabling the smooth transfer from one to the other via the α-divergence. We reveal that generalizations of the α-Bridge are closely related to approaches developed recently to regularize adversarial learning, providing insights into that prior work, and further understanding of why the α-Bridge performs well in practice.

Duke Scholars

Published In

AAAI 2020 - 34th AAAI Conference on Artificial Intelligence

Publication Date

January 1, 2020

Start / End Page

6901 / 6908
 

Citation

APA
Chicago
ICMJE
MLA
NLM
Zhao, M., Cong, Y., Dai, S., & Carin, L. (2020). Bridging maximum likelihood and adversarial learning via α-divergence. In AAAI 2020 - 34th AAAI Conference on Artificial Intelligence (pp. 6901–6908).
Zhao, M., Y. Cong, S. Dai, and L. Carin. “Bridging maximum likelihood and adversarial learning via α-divergence.” In AAAI 2020 - 34th AAAI Conference on Artificial Intelligence, 6901–8, 2020.
Zhao M, Cong Y, Dai S, Carin L. Bridging maximum likelihood and adversarial learning via α-divergence. In: AAAI 2020 - 34th AAAI Conference on Artificial Intelligence. 2020. p. 6901–8.
Zhao, M., et al. “Bridging maximum likelihood and adversarial learning via α-divergence.” AAAI 2020 - 34th AAAI Conference on Artificial Intelligence, 2020, pp. 6901–08.
Zhao M, Cong Y, Dai S, Carin L. Bridging maximum likelihood and adversarial learning via α-divergence. AAAI 2020 - 34th AAAI Conference on Artificial Intelligence. 2020. p. 6901–6908.

Published In

AAAI 2020 - 34th AAAI Conference on Artificial Intelligence

Publication Date

January 1, 2020

Start / End Page

6901 / 6908