Timezone: »
The instability in GANs' training has been a long-standing problem despite remarkable research efforts. We identify that instability issues stem from difficulties of performing feature matching with mini-batch statistics, due to a fragile balance between the fixed target distribution and the progressively generated distribution. In this work, we propose feature quantizatoin (FQ) for the discriminator, to embed both true and fake data samples into a shared discrete space. The quantized values of FQ are constructed as an evolving dictionary, which is consistent with feature statistics of the recent distribution history. Hence, FQ implicitly enables robust feature matching in a compact space. Our method can be easily plugged into existing GAN models, with little computational overhead in training. Extensive experimental results show that the proposed FQ-GAN can improve the FID scores of baseline methods by a large margin on a variety of tasks, including three representative GAN models on 10 benchmarks, achieving new state-of-the-art performance.
Author Information
Yang Zhao (University at Buffalo)
Chunyuan Li (Microsoft Research)
Ping Yu (Sony Interactive Entertainment LLC)
Jianfeng Gao (Microsoft Research AI)
Jianfeng Gao is Partner Research Manager at Microsoft Research AI. He leads the development of AI systems for machine reading comprehension (MRC), question answering (QA), social bots, goal-oriented dialogue, and business applications. From 2014 to 2017, he was Partner Research Manager at Deep Learning Technology Center at Microsoft Research, Redmond, where he was leading the research on deep learning for text and image processing. From 2006 to 2014, he was Principal Researcher at Natural Language Processing Group at Microsoft Research, Redmond, where he worked on Web search, query understanding and reformulation, ads prediction, and statistical machine translation. From 2005 to 2006, he was a Research Lead in Natural Interactive Services Division at Microsoft, where he worked on Project X, an effort of developing natural user interface for Windows. From 2000 to 2005, he was Research Lead in Natural Language Computing Group at Microsoft Research Asia, where he and his colleagues developed the first Chinese speech recognition system released with Microsoft Office, the Chinese/Japanese Input Method Editors (IME) which were the leading products in the market, and the natural language platform for Microsoft Windows.
Changyou Chen (SUNY Buffalo)
More from the Same Authors
-
2020 Poster: UniLMv2: Pseudo-Masked Language Models for Unified Language Model Pre-Training »
Hangbo Bao · Li Dong · Furu Wei · Wenhui Wang · Nan Yang · Xiaodong Liu · Yu Wang · Jianfeng Gao · Songhao Piao · Ming Zhou · Hsiao-Wuen Hon -
2020 Poster: Mapping natural-language problems to formal-language solutions using structured neural representations »
Kezhen Chen · Qiuyuan Huang · Hamid Palangi · Paul Smolensky · Ken Forbus · Jianfeng Gao -
2020 Poster: Variance Reduction in Stochastic Particle-Optimization Sampling »
Jianyi Zhang · Yang Zhao · Changyou Chen -
2019 Poster: Differentially Private Empirical Risk Minimization with Non-convex Loss Functions »
Di Wang · Changyou Chen · Jinhui Xu -
2019 Oral: Differentially Private Empirical Risk Minimization with Non-convex Loss Functions »
Di Wang · Changyou Chen · Jinhui Xu -
2019 Tutorial: Neural Approaches to Conversational AI »
Michel Galley · Jianfeng Gao