Challenges

Description:

Procedural content generation (PCG) refers to the practice of generating game content, such as levels, quests, or characters, algorithmically. In this challenge, you are asked to generate a playable, and coherent level for Super Mario Bros. using the latest PCG techniques.

Paper:

TOAD-GAN: Coherent Style Level Generation from a Single Example

Description:

Object recognition is a computer vision technique for identifying objects in images or videos. Object recognition is a key output of deep learning and machine learning algorithms. In this challenge, you are asked to apply spiking neural networks to accomplish the purpose.

Paper:

1. https://www.sciencedirect.com/science/article/abs/pii/S0893608017302903

2. https://www.sciencedirect.com/science/article/abs/pii/S0893608018303332

3. https://www.frontiersin.org/articles/10.3389/fnins.2018.00774/full

4. https://www.frontiersin.org/articles/10.3389/fnins.2017.00682/full

Description:

The purpose of a brain-computer interface (BCI) is to have a direct communication pathway between the brain and an external device. That enables its users to interact with computers by means of brain activity. a dataset will be given to you for purpose of decoding the brain. the dataset contains EEG signals related to motor imagery tasks. you are asked to classify them.

Paper:

1. https://onlinelibrary.wiley.com/doi/full/10.1002/hbm.23730

2. https://iopscience.iop.org/article/10.1088/1741-2552/ab0ab5/meta

Description:

Deep reinforcement learning is a subfield of machine learning that combines reinforcement learning and deep learning. RL considers the problem of a computational agent learning to make decisions by trial and error. You are free to choose between available methods. The final assessment will be based on the score obtained by your agent and the novelty of your work.

Paper:

1. https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf

2. https://arxiv.org/pdf/1710.02298.pdf

3. https://arxiv.org/pdf/1602.01783.pdf

Description:

A generative adversarial network (GAN) is a class of machine learning frameworks designed by Ian Goodfellow and his colleagues in 2014. Two neural networks contesting with each other in a game (in the form of a zero-sum game, where one agent's gain is another agent's loss). you are asked to train a GAN which can generate an image from a given input image.

Paper:

1. https://arxiv.org/abs/1611.07004v3

2. https://arxiv.org/abs/1406.2661

Description:

Emotion recognition is the process of identifying human emotion. People vary widely in their accuracy at recognizing the emotions of others. The Use of technology to help people with emotion recognition is a relatively nascent research area. Competitors are required to perform the emotion recognition task on the ShEMO dataset.

Paper:

https://arxiv.org/abs/1906.01155

Description:

Language modeling (LM) is the use of various statistical and probabilistic techniques to determine the probability of a given sequence of words occurring in a sentence. Language models analyze bodies of text data to provide a basis for their word predictions. Participants are given the Bijankhan corpus and asked to produce a language model which is evaluated based on their perplexity.

Paper:

1. https://arxiv.org/abs/1706.03762

2. https://arxiv.org/abs/1810.04805

3. https://arxiv.org/abs/2005.12515

Description:

Time series analysis is a statistical technique that deals with time series data, or trend analysis. Time series data means that data is in a series of particular time periods or intervals. Historical data of S&P 500 will be given to competitors and they will be asked to forecast the future of the index.

Paper:

1. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0227222