Data Phoenix Digest - ISSUE 57

Data Phoenix Events

Data Phoenix team invites you all to our upcoming "The A-Z of Data" charity webinar that’s going to take place on November 16, 2022 at 16.00 CET.

• Topic: “Natural Language Processing (NLP) and Machine Learning in Healthcare”.
• Speaker: Dr. Hemachandran Kannan, Zita Zoltay Paprika Professor of Decision Sciences & Business Economics and Course5i Chair Professor of Business Analytics and Machine Learning at Woxsen University
• Language: English
• Participation: free (but you’ll be required to register)
• Karma perk: donate to our charity initiative

NEWS

• Gen AI Market Map by Sequoia Capital Venture Investors
• Red Hat and IBM tandem to improve AIops with open source project
• Getting to know IBM's artificial intelligence unit
• Combining fitness trackers and artificial intelligence to detect depression
• Deep Learning with Light

PAPERS

YOWO-Plus: An Incremental Improvement

Developers make updates to the YOWO design to make it better. For the network structure, they use the same ones of official implemented YOWO, including 3D-ResNext-101 and the best pre-trained weight of the re-implemented YOLOv2, which is better than the official YOLOv2.

AutoAvatar: Autoregressive Neural Fields for Dynamic Avatar Modeling

AutoAvatar is an autoregressive approach for modeling dynamically deforming human bodies directly from raw scans.

High Fidelity Neural Audio Compression

Meta Fundamental AI Research (FAIR) team on audio hypercompression shows how AI can be used to ensure that audio messages don't glitch or slow down when the Internet connection is poor.

DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation
Large text-to-image models lack the ability to mimic the appearance of subjects and synthesize novel renditions of them. This paper presents a new approach for "personalization" of text-to-image diffusion models.

TEACH: Temporal Action Compositions for 3D Humans
The authors demonstrate how they can generate 3D human motions that correspond semantically to the text, and follow the temporal order of the instructions, to enable the synthesis of a series of actions.

Content-Based Search for Deep Generative Models
Modelverse is a model sharing platform that contains a diverse set of deep generative models, such as animals, landscapes, portraits, and art pieces. Learn more about its a content-based search engine.

VToonify: Controllable High-Resolution Portrait Video Style Transfer
In this paper, the authors investigate the challenges of controllable high-resolution portrait video style transfer by introducing a novel VToonify framework. Check it out — it’s amazing!

MetaFormer Baselines for Vision
This paper explores the capacity of MetaFormer and introduces several baseline models under MetaFormer using the most basic or common mixers. Check out the authors’ observations.

ARTICLES

Apply 100 ML Models with Hyperparameter Tuning Using 3 Lines of Code
Auto-sklearn library enables you to apply bulk ML Models to datasets, yet it doesn’t support Windows. Find out how to use Google Colab, Kaggle, or Docker to install the auto-sklearn library.

Outperform OpenAI GPT-3 with SetFit for text-classification
In this article, you’ll learn how to use SetFit to create an efficient text-classification model. You will also learn how you can improve your model by using hyperparameter tuning.

5 Tools That Will Help You Setup Production ML Model Testing
In this article, you’ll learn about the tools that can be used to test an ML model: both open-source and paid. This article fully explores the tools for implementing your MLOps pipeline.

I made an AI that can study for me.
The authors explains how he created an AI helper that can study the content (various texts) for him and answer the questions he directly asks it using the roBERTa model and MLM.

VIDEOS

Speed Up Presto at Uber with Alluxio Caching
Presto is a primary data analytics tool at Uber. Learn how Uber integrated Alluxio, an open-source data orchestration platform, into Presto to achieve performance improvements in caching.

We hope that you liked the digest. Kindly help us make DataPhoenix a better place for all readers. Please, take part in our survey — It won’t take more than a few minutes!