Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. A Better Use of Audio-Visual Cues: Dense Video Captioning with Bi-modal Transformer (Tampere University, Finland, 2020)
2. Fast Bi-layer Neural Synthesis of One-Shot Realistic Head Avatars (Samsung AI Center, 2020)
3. Enhancing the Locality and Breaking the Memory Bottleneck of Transformer on Time Series Forecasting (University of California, USA, 2019)
4. Whitening for Self-Supervised Representation Learning (University of Trento, Italy, 2020)
5. MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesis (Lyrebird AI and University of Montreal, 2019)
6. StyleFlow: Attribute-conditioned Exploration of StyleGAN-Generated Images using Conditional Continuous Normalizing Flows (KAUST, Adobe, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. High-Resolution Neural Face Swapping for Visual Effects (Disney Research Studios, ETH Zurich, 2020)
2. Beyond Accuracy: Behavioral Testing of NLP Models with CheckList (USA, 2020)
3. Thieves on Sesame Street! Model Extraction of BERT-based APIs (UMass & Google Research, ICLR, 2019)
4. Time-Aware User Embeddings as a Service (Yahoo! Research, Temple University, 2020)
5. Are Labels Necessary for Neural Architecture Search? (Johns Hopkins University, Facebook AI Research, 2020)
6. GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding (Google, 2020)
7. Data Shapley: Equitable Valuation of Data for Machine Learning (USA, 2019)
8. Language-agnostic BERT Sentence Embedding (Google AI, 2020)
9. Self-Supervised Learning for Large-Scale Unsupervised Image Clustering (Technion, Israel, 2020)
10. Batch-Channel Normalization and Weight Standardization (2 papers, Johns HopkinsUniversity, USA, 2019)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. PointRend: Image Segmentation as Rendering (Facebook AI Research, 2020)
2. Natural- To Formal-Language Generation Using Tensor Product Representations (USA, 2019)
3. Linformer: Self-Attention with Linear Complexity (Facebook AI, 2020)
4. DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution (Johns Hopkins University, Google, 2020)
5. Training Generative Adversarial Networks with Limited Data (NVIDIA, 2020)
6. Multi-Modal Dense Video Captioning (Tampere University, Finland, 2020
7. Are we done with ImageNet? (DeepMind, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks (China, 2020)
2. TAPAS: Weakly Supervised Table Parsing via Pre-training (Google, 2020)
3. DeepFaceLab: A simple, flexible and extensible faceswapping framework (2020)
4. End-to-End Object Detection with Transformers (Facebook AI, 2020)
5. Language Models are Few-Shot Learners (OpenAI, 2020)
6. TabNet: Attentive Interpretable Tabular Learning (Google Cloud AI, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Efficient Document Re-Ranking for Transformers by Precomputing Term Representations; EARL: Speedup Transformer-based Rankers with Pre-computed Representation (2020)
2. MakeItTalk: Speaker-Aware Talking Head Animation (Adobe, University of Massachusetts Amherst, Huya, 2020)
3. Jukebox: A Generative Model for Music (OpenAI, 2020)
4. Recipes for building an open-domain chatbot (Facebook AI Research, 2020)
5. One-Shot Object Detection without Fine-Tuning (HKUST, Hong Kong, Tencent, 2020)
6. f-BRS: Rethinking Backpropagating Refinement for Interactive Segmentation (Samsung AI Center, Moscow, 2020)
7. Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis (NVIDIA, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization (Georgia Institute of Technology, Atlanta, USA, 2016)
2. X3D: Expanding Architectures for Efficient Video Recognition (Facebook AI Research, 2020)
3. Adaptive Attention Span in Transformers (Facebook AI Research, 2019)
4. ResNeSt: Split-Attention Networks (Amazon, 2020)
5. Weight Standardization (Johns Hopkins University, 2019)
6. Supervised Contrastive Learning (Google Research, MIT, 2020)
7. Improved Training Speed, Accuracy, and Data Utilization Through Loss Function Optimization (USA, 2019)
8. TTNet: Real-time temporal and spatial video analysis of table tennis (OSAI, 2020)
9. Learning in the Frequency Domain (Alibaba, Arizona, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. TResNet: High Performance GPU-Dedicated Architecture (DAMO Academy, Alibaba Group, 2020)
2. Controllable Person Image Synthesis with Attribute-Decomposed GAN (China, 2020)
3. Learning to See Through Obstructions (Taiwan, USA, 2020)
4. Tracking Objects as Points (UT Austin, Intel Labs, 2020)
5. CookGAN: Meal Image Synthesis from Ingredients (USA, UK, 2020)
6. Designing Network Design Spaces (FAIR, 2020)
7. Gradient Centralization: A New Optimization Technique for Deep Neural Networks (Hong Kong, Alibaba, 2020)
8. When Does Unsupervised Machine Translation Work? (Johns Hopkins University, USA, 2020)

Привет, Хабр!

Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество! Первая часть мартовской сборки обзоров опубликована ранее.

Статьи на сегодня:

1. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis (UC Berkeley, Google Research, UC San Diego, 2020)
2. Scene Text Recognition via Transformer (China, 2020)
3. PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization (Imperial College London, Google Research, 2019)
4. Lagrangian Neural Networks (Princeton, Oregon, Google, Flatiron, 2020)
5. Deformable Style Transfer (Chicago, USA, 2020)
6. Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need? (MIT, Google, 2020)
7. Attentive CutMix: An Enhanced Data Augmentation Approach for Deep Learning Based Image Classification (Carnegie Mellon University, USA, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Fast Differentiable Sorting and Ranking (Google Brain, 2020)
2. MaxUp: A Simple Way to Improve Generalization of Neural Network Training (UT Austin, 2020)
3. Deep Nearest Neighbor Anomaly Detection (Jerusalem, Israel, 2020)
4. AutoML-Zero: Evolving Machine Learning Algorithms From Scratch (Google, 2020)
5. SpERT: Span-based Joint Entity and Relation Extraction with Transformer Pre-training (RheinMain University, Germany, 2019)
6. High-Resolution Daytime Translation Without Domain Labels (Samsung AI Center, Moscow, 2020)
7. Incremental Few-Shot Object Detection (UK, 2020)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Представлены обзоры 11 статей по Computer Vision, Natural Language Processing, Reinforcement learning и другим темам.

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Poly-encoders: Transformer Architectures and Pre-training Strategies for Fast and Accurate Multi-sentence Scoring (Facebook, 2019)
2. Implicit Discriminator in Variational Autoencoder (Indian Institute of Technology Ropar, 2019)
3. Self-training with Noisy Student improves ImageNet classification (Google Research, Carnegie Mellon University, 2019)
4. Momentum Contrast for Unsupervised Visual Representation Learning (Facebook, 2019)
5. Benchmarking Neural Network Robustness to Common Corruptions and Perturbations (University of California, Oregon State University, 2019)
6. DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter (Hugging Face, 2019)
7. Plug and Play Language Models: A Simple Approach To Controlled Text Generation (Uber AI, Caltech, HKUST, 2019)
8. Deep Salience Representation for F0 Estimation in Polyphonic Music ( New York University, USA, 2017)
9. Analyzing and Improving the Image Quality of StyleGAN (NVIDIA, 2019)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Layer rotation: a surprisingly powerful indicator of generalization in deep networks? (Université catholique de Louvain, Belgium, 2018)
2. Parameter-Efficient Transfer Learning for NLP (Google Research, Jagiellonian University, 2019)
3. RoBERTa: A Robustly Optimized BERT Pretraining Approach (University of Washington, Facebook AI, 2019)
4. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (Google Research, 2019)
5. How the Brain Transitions from Conscious to Subliminal Perception (USA, Argentina, Spain, 2019)
6. Large Memory Layers with Product Keys (Facebook AI Research, 2019)
7. Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches (Politecnico di Milano, University of Klagenfurt, 2019)
8. Omni-Scale Feature Learning for Person Re-Identification (University of Surrey, Queen Mary University, Samsung AI, 2019)
9. Neural reparameterization improves structural optimization (Google Research, 2019)

Привет, Хабр! Продолжаем публиковать рецензии на научные статьи от членов сообщества Open Data Science из канала #article_essense. Хотите получать их раньше всех — вступайте в сообщество!

Статьи на сегодня:

1. Neural Ordinary Differential Equations (University of Toronto, 2018)
2. Semi-Unsupervised Learning with Deep Generative Models: Clustering and Classifying using Ultra-Sparse Labels (University of Oxford, The Alan Turing Institute, London, 2019)
3. Uncovering and Mitigating Algorithmic Bias through Learned Latent Structure (Massachusetts Institute of Technology, Harvard University, 2019)
4. Deep reinforcement learning from human preferences (OpenAI, DeepMind, 2017)
5. Exploring Randomly Wired Neural Networks for Image Recognition (Facebook AI Research, 2019)
6. Photofeeler-D3: A Neural Network with Voter Modeling for Dating Photo Rating (Photofeeler Inc., 2019)
7. MixMatch: A Holistic Approach to Semi-Supervised Learning (Google Reasearch, 2019)
8. Divide and Conquer the Embedding Space for Metric Learning (Heidelberg University, 2019)