Learning Generative Adversarial Networks from Multiple Data Sources

Le, Trung; Hoang, Quan; Vu, Hung; Nguyen, Tu Dinh; Bui, Hung; Phung, Dinh

doi:10.24963/ijcai.2019/391

Cited by 8 publications

(3 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MEGAN [31] adopts a gating network that produces a one-hot vector to select the generator creating the best example. P2GAN [32] sequentially adds a new generator, which is different from the existing generators, to cover the missing modes of the real data.…”

Section: Gan With Multiple Generatorsmentioning

confidence: 99%

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

Choi¹,

Han²

2021

Preprint

View full text Add to dashboard Cite

We propose a generative adversarial network with multiple discriminators, where each discriminator is specialized to distinguish the subset of a real dataset. This approach facilitates learning a generator coinciding with the underlying data distribution and thus mitigates the chronic mode collapse problem. From the inspiration of multiple choice learning, we guide each discriminator to have expertise in the subset of the entire data and allow the generator to find reasonable correspondences between the latent and real data spaces automatically without supervision for training examples and the number of discriminators. Despite the use of multiple discriminators, the backbone networks are shared across the discriminators and the increase of training cost is minimized. We demonstrate the effectiveness of our algorithm in the standard datasets using multiple evaluation metrics.Preprint. Under review.

show abstract

Section: Gan With Multiple Generatorsmentioning

confidence: 99%

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

Choi¹,

Han²

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Unlike DAgger (Ross, Gordon, and Bagnell 2011), can simply ask the expert for such actions. GAIL suffers from the problems of mode collapse and low sample efficiency in terms of environment interaction (Le et al 2019). The weakness of mode collapse is inheriting from GANs, and several works have built on GAIL to overcome this problem (Li, Song, and Ermon 2017) (Fei et al 2020).…”

Section: Introductionmentioning

confidence: 99%

Sample Efficient Imitation Learning via Reward Function Trained in Advance

Zhang¹

2021

Preprint

View full text Add to dashboard Cite

Imitation learning (IL) is a framework that learns to imitate expert behavior from demonstrations. Recently, IL shows promising results on high dimensional and control tasks. However, IL typically suffers from sample inefficiency in terms of environment interaction, which severely limits their application to simulated domains. In industrial applications, learner usually have a high interaction cost, the more interactions with environment, the more damage it causes to the environment and the learner itself. In this article, we make an effort to improve sample efficiency by introducing a novel scheme of inverse reinforcement learning. Our method, which we call Model Reward Function Based Imitation Learning (MRFIL), uses an ensemble dynamic model as a reward function, what is trained with expert demonstrations. The key idea is to provide the agent with an incentive to match the demonstrations over a long horizon, by providing a positive reward upon encountering states in line with the expert demonstration distribution. In addition, we demonstrate the convergence guarantee for new objective function. Experimental results show that our algorithm reaches the competitive performance and significantly reducing the environment interactions compared to IL methods.

show abstract

“…GANs are known to be affected by the mode collapsing problem [5,7,10,17]. In particular, the study in [17] recently studied the mode collapsing problem and further classified this into the missing mode problem i.e., the generated samples miss some modes in the true data, and the boundary distortion problem i.e., the generated samples can only partly recover some modes in the true data.…”

Section: Introductionmentioning

confidence: 99%

Dual-Component Deep Domain Adaptation: A New Approach for Cross Project Software Vulnerability Detection

Nguyen

Vel

et al. 2020

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Owing to the ubiquity of computer software, software vulnerability detection (SVD) has become an important problem in the software industry and computer security. One of the most crucial issues in SVD is coping with the scarcity of labeled vulnerabilities in projects that require the laborious manual labeling of code by software security experts. One possible solution is to employ deep domain adaptation (DA) which has recently witnessed enormous success in transferring learning from structural labeled to unlabeled data sources. Generative adversarial network (GAN) is a technique that attempts to bridge the gap between source and target data in the joint space and emerges as a building block to develop deep DA approaches with state-of-the-art performance. However, deep DA approaches using the GAN principle to close the gap are subject to the mode collapsing problem that negatively impacts the predictive performance. Our aim in this paper is to propose Dual Generator-Discriminator Deep Code Domain Adaptation Network (Dual-GD-DDAN) for tackling the problem of transfer learning from labeled to unlabeled software projects in SVD to resolve the mode collapsing problem faced in previous approaches. The experimental results on real-world software projects show that our method outperforms stateof-the-art baselines by a wide margin.

show abstract

Learning Generative Adversarial Networks from Multiple Data Sources

Cited by 8 publications

References 5 publications

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators

Sample Efficient Imitation Learning via Reward Function Trained in Advance

Dual-Component Deep Domain Adaptation: A New Approach for Cross Project Software Vulnerability Detection

Contact Info

Product

Resources

About