InfoQ Homepage Model Content on InfoQ
-
The Unreasonable Effectiveness of Zero Shot Learning
Roland Meertens shows how one can get started deploying models without requiring any data, discussing foundational models, and examples of them, such as GPT-3 and OpenAI CLIP.
/presentations/deploing-models-gpt3-openai-clip/en/mediumimage/roland-meertens-m-1655444707687.jpg)
-
Unified MLOps: Feature Stores and Model Deployment
Monte Zweben proposes a whole new approach to MLOps that allows to scale models without increasing latency by merging a database, a feature store, and machine learning.
/presentations/mlops-model-deployment/en/mediumimage/monte_zweben-m-1652696083026.jpg)
-
Iterating on Models on Operating ML
Monte Zweben and Roland Meertens discuss the challenges in building, maintaining, and operating machine learning models.
/presentations/operating-ml-models/en/mediumimage/Untitled-1-1650979197460.jpg)
-
Deep Learning at Scale: Distributed Training and Hyperparameter Search for Image Recognition Problems
Michael Shtelma discusses methods and libraries for training models on a dataset that does not fit into memory or maybe even on the disk using multiple GPUs or even nodes.
/presentations/train-large-models/en/mediumimage/Micbig-1590503376615.JPG)
-
From Spark to Elasticsearch and Back - Learning Large-Scale Models for Content Recommendation
Sonya Liberman shares an algorithmic architecture that enables running complex models under difficult scale constraints and shortens the cycle between research and production.
/presentations/content-recommendation-model/en/mediumimage/Sonbig-1584280027769.JPG)
-
ML's Hidden Tasks: A Checklist for Developers When Building ML Systems
Jade Abbott discusses the set of unexpected things that go on the "take it to production" checklist in the case of machine learning, and what are the tools that can help.
/presentations/ml-hidden-tasks/en/mediumimage/Jade-Abbott-m-1580858042834.jpg)
-
A Look at the Methods to Detect and Try to Remove Bias in Machine Learning Models
Thierry Silbermann explores some examples where machine learning fails and/or is making a negative impact, looking at some of the tools available today to fix the model.
/presentations/bias-ml-sao-paulo-2019/en/mediumimage/Tibig-1569540287492.jpg)
-
Deep Learning for Recommender Systems
Oliver Gindele discusses how some DL models can be implemented in TensorFlow, starting from a collaborative filtering approach and extending that to more complex deep recommender systems.
/presentations/dl-models-tensorflow/en/mediumimage/Oliver-Gindele-m-1564599110695.jpg)
-
Petastorm: A Light-Weight Approach to Building ML Pipelines
Yevgeni Litvin describes how Petastorm facilitates tighter integration between Big Data and Deep Learning worlds, simplifies data management and data pipelines, and speeds up model experimentation.
/presentations/petastorm-ml-pipelines/en/mediumimage/Yevgeni-Litvin-m-1560415871968.jpg)
-
Ludwig: A Code-Free Deep Learning Toolbox
Piero Molino introduces Ludwig, a deep learning toolbox that allows to train models and to use them for prediction without the need to write code.
/presentations/ludwig-dl-toolbox/en/mediumimage/piero_molino-h-1559952531778.jpeg)
-
BDD and the New Model for Testing
Paul Gerrard proposes a model of the thought processes that every tester uses which maps directly to the BDD way, helping practitioners understand the BDD collaboration and test process.
/presentations/bdd-model-testing/en/mediumimage/Paul-Gerrard-m-1556953614785.jpg)
-
Test-Driven Machine Learning
Detlef Nauck explains why the testing of data is essential, as it not only drives the machine learning phase itself, but it is paramount for producing reliable predictions after deployment.
/presentations/tdd-ml/en/mediumimage/Qcon-London-Detlef-Nauck-m-1556177499049.jpg)