Exploring the evolution of GANs through quality diversity

Costa, Victor; Lourenço, Nuno; Correia, João; Machado, Penousal

doi:10.1145/3377930.3389824

Cited by 14 publications

(7 citation statements)

References 19 publications

(43 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, they are implicitly parallel, which is suitable for execution on massive clusters and allows a joint exploration of the search space, offering greater resilience over the highly non-convex multi-modal nature of many real-life problems. In particular, the recent trend of quality-diversity optimization algorithms [9,25] indicates the importance of different individuals during the optimization process. In this work, we are using population-based algorithms given their capacity to adapt to a given context, which is required to generate effective and robust general optimizers.…”

Section: Related Workmentioning

confidence: 99%

“…More specifically, we consider the invariance to monotonically increasing transformations of 𝑓 . For example, the performance of the optimizer in 𝑓 is the same in 𝑓 3 , 𝑓 × 2 |𝑓 | −5 9 , etc. Therefore, we replace all of the evaluations on 𝑓 by an adaptive transformation of 𝑓 to represent how the observed values are relative to other observations in the current step [28].…”

Section: Experimental and Implementation Setupmentioning

confidence: 99%

See 1 more Smart Citation

Meta Learning Black-Box Population-Based Optimizers

Gomes,

Léger,

Gagné

2021

Preprint

View full text Add to dashboard Cite

The no free lunch theorem states that no model is better suited to every problem. A question that arises from this is how to design methods that propose optimizers tailored to specific problems achieving state-of-the-art performance. This paper addresses this issue by proposing the use of meta-learning to infer population-based blackbox optimizers that can automatically adapt to specific classes of problems. We suggest a general modeling of population-based algorithms that result in Learning-to-Optimize POMDP (LTO-POMDP), a meta-learning framework based on a specific partially observable Markov decision process (POMDP). From that framework's formulation, we propose to parameterize the algorithm using deep recurrent neural networks and use a meta-loss function based on stochastic algorithms' performance to train efficient data-driven optimizers over several related optimization tasks. The learned optimizers' performance based on this implementation is assessed on various black-box optimization tasks and hyperparameter tuning of machine learning models. Our results revealed that the meta-loss function encourages a learned algorithm to alter its search behavior so that it can easily fit into a new context. Thus, it allows better generalization and higher sample efficiency than state-of-the-art generic optimization algorithms, such as the Covariance matrix adaptation evolution strategy (CMA-ES).

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Experimental and Implementation Setupmentioning

confidence: 99%

Meta Learning Black-Box Population-Based Optimizers

Gomes,

Léger,

Gagné

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Mustangs is inspired by E-GAN and applies different loss functions to competitive co-evolutionary algorithm [12]. In addition, some researchers combine GANs with evolutionary strategies in their own ways [13][14][15][16][17]. Garciarena et al use crossover in their research [17], which is as common in biology as mutation.…”

Section: Introductionmentioning

confidence: 99%

Evolutionary Generative Adversarial Networks with Crossover Based Knowledge Distillation

Li¹,

Zhang²,

Gong³

et al. 2021

Preprint

View full text Add to dashboard Cite

Generative Adversarial Networks (GAN) is an adversarial model, and it has been demonstrated to be effective for various generative tasks. However, GAN and its variants also suffer from many training problems, such as mode collapse and gradient vanish. In this paper, we firstly propose a general crossover operator, which can be widely applied to GANs using evolutionary strategies. Then we design an evolutionary GAN framework C-GAN based on it. And we combine the crossover operator with evolutionary generative adversarial networks (E-GAN) to implement the evolutionary generative adversarial networks with crossover (CE-GAN). Under the premise that a variety of loss functions are used as mutation operators to generate mutation individuals, we evaluate the generated samples and allow the mutation individuals to learn experiences from the output in a knowledge distillation manner, imitating the best output outcome, resulting in better offspring. Then, we greedily selected the best offspring as parents for subsequent training using discriminator as evaluator. Experiments on real datasets demonstrate the effectiveness of CE-GAN and show that our method is competitive in terms of generated images quality and time efficiency.

show abstract

“…With this in mind, to evaluate the generated output of a GAN, researchers typically focus on aspects including, similarity, diversity, and plausibility [5,6,7]. Currently, for generated audio, time-consuming human listening studies are a common qualitative approach to observe these aspects.…”

Section: Introductionmentioning

confidence: 99%

A Prototypical Network Approach for Evaluating Generated Emotional Speech

Baird¹,

Mertes²,

Milling³

et al. 2021

Interspeech 2021

View full text Add to dashboard Cite

The collection of emotional speech data is a time-consuming and costly endeavour. Generative networks can be applied to augment the limited audio data artificially. However, it is challenging to evaluate generated audio for its similarity to source data, as current quantitative metrics are not necessarily suited to the audio domain. We explore the use of a prototypical network to evaluate four classes of generated emotional audio with this in mind. We first extract spectrogram images from WAVEGAN generated audio and other audio augmentation approaches, comparing similarity to the class prototype and diversity within the embedding space. Furthermore, we augment the source training set with each augmentation type and perform a classification to explore the generated audio plausibility. Results suggest that quality and diversity can be quantitatively observed with this approach. In the chosen context, we see that WAVEGAN generated data is recognisable as a source data class (F1-score 43.6 %), and the samples add similar diversity as unseen source data. This result leads to more plausible data for augmentation of the source training set -achieving up to 63.9 % F1 which is a 3.5 % improvement over the source data baseline.

show abstract

Exploring the evolution of GANs through quality diversity

Cited by 14 publications

References 19 publications

Meta Learning Black-Box Population-Based Optimizers

Meta Learning Black-Box Population-Based Optimizers

Evolutionary Generative Adversarial Networks with Crossover Based Knowledge Distillation

A Prototypical Network Approach for Evaluating Generated Emotional Speech

Contact Info

Product

Resources

About