Published inTowards Data ScienceThe Evolved Transformer — Enhancing Transformer with Neural Architecture SearchNeural architecture search (NAS) is the process of algorithmically searching for new designs of neural networks. Though researchers have…Mar 14, 20192Mar 14, 20192
Published inTowards Data ScienceXLM — Enhancing BERT for Cross-lingual Language ModelCross-lingual Language Model PretrainingFeb 12, 20195Feb 12, 20195
Published inTowards Data ScienceIdentifying and Correcting Label Bias in Machine LearningIn a new paper from Google, researchers propose a novel technique to train machine learning algorithms fairly even with a biased dataset.Feb 9, 20192Feb 9, 20192
Published inTowards Data ScienceTransformer-XL Explained: Combining Transformers and RNNs into a State-of-the-art Language ModelLanguage modeling has become an important NLP technique thanks to the ability to apply it to various NLP tasks, such as machine…Jan 17, 20192Jan 17, 20192
Published inTowards Data ScienceExplained: Multilingual Sentence Embeddings for Zero-Shot TransferApplying a Single Model on 93 LanguagesJan 7, 20192Jan 7, 20192
Published inSnipSnip — December Update & 2018 SummaryGreetings to the Snip community!Jan 3, 2019Jan 3, 2019
Published inTowards Data ScienceExplained: A Style-Based Generator Architecture for GANs - Generating and Tuning Realistic…NVIDIA’s novel architecture for Generative Adversarial NetworksDec 30, 20189Dec 30, 20189
Published inTowards Data ScienceSlowFast Explained - Dual-mode CNN for Video UnderstandingState-of-the-art Deep Learning architecture inspired by the visual mechanism of primatesDec 27, 20185Dec 27, 20185
Published inTowards Data ScienceKuzushiji-MNIST - Japanese Literature Alternative Dataset for Deep Learning TasksPlus our VGG-ResNet ensemble model with state-of-the-art resultsDec 14, 2018Dec 14, 2018
