ADVERSARIAL MASK TRANSFORMER FOR SEQUENTIAL LEARNING

Keyphrases

• Attention Layer 33%
• Contextual Embeddings 33%
• Downstream Applications 33%
• Downstream Task 33%
• Language Comprehension 66%
• Language Model 33%
• Low-resource 33%
• Machine Translation 33%
• Mask Mechanism 33%
• Mask Transformers 100%
• Masked Language Model 100%
• Minimax Optimization 33%
• Pre-training Tasks 33%
• Prediction Loss 33%
• Random Mask 33%
• Sequence Learning 33%
• Sequential Learning 100%
• Training Model 33%
• Transformer 33%
• Transformer-based Language Models 33%
• Translation Task 33%
• Two-machine 33%
• Word Prediction 66%

Computer Science

• Attention (Machine Learning) 20%
• Language Modeling 100%
• Language Understanding 40%
• Machine Translation 20%
• Pre-Trained Model 20%
• Seq2Seq 20%
• Sequential Learning 100%

Psychology

• Sequence Learning 100%