Generative Adversarial Network
Generative Adversarial Network modeling (GAN) involves using a model to generate new examples that plausibly come from an existing distribution of samples, many application of GAN such as generate examples for image datasets, image-to-image translation, text-to-image translation, face frontal view generation, and so on. GAN is still potential topics in machine learning in general.
Deep learning techniques show dramatic performance on computer vision area, and they even outperform human. But it is also vulnerable to some small perturbation called an adversarial attack. This is a problem combined with the safety of artificial intelligence. These attacks have shown that they can fool models of image classification, semantic segmentation, and object detection.
On this topic, we have Seungju, Mingu who're diligent working and have some good published works and still dive deep into it. We're willing to share and help you if you're interested in.
Deep learning on graph data is tough but very potential topic recently. We're currently collaborating and working on it.
Continual learning (CL) is the ability of a model to learn continually from a stream of data, building on what was learnt previously, hence exhibiting positive transfer, as well as being able to remember previously seen tasks.
In this task, we have Giang who eagerly researched and proposed potential works. We're willing to share and help.
Artificial production of human speech is known as speech synthesis. This machine learning-based technique is applicable in text-to-speech, music generation, speech generation, speech-enabled devices, navigation systems, and accessibility for visually-impaired people.
Currently we have Keon are keenly working on with potential ideas and willing to share.