Blog

    Posts about Data Science:

    filter icon
    Filter
    Model remediation before and after

    Model Remediation: The Solution to The Day 3 Problem of Model Drift

    Model failure is a huge concern for any AI-driven organization. The phenomenon that causes models to stop working in production—known as model drift,...
    Read article
    Cartoon of Striveworks CSO Eric Korman drinking coffee

    What Is Valor, the Striveworks Model Evaluation Service? [Video]

    Valor is a first-of-its-kind AI model evaluation service from Striveworks. This open-source tool saves data teams enormous time and effort in...
    Read article
    Three grids, each showing a two data distributions progressively farther and farther apart.

    What Is AI Model Drift?

    In our recent post on model retraining, we touch on an unfortunate but unavoidable fact of machine learning (ML) models: They often have remarkably...
    Read article
    Floating icons and user cards related to choosing an MLOps vendor connected in a web

    How to Choose an MLOps Vendor

    Once you’ve settled the “build vs. buy” debate, your research enters a new phase: how to choose a machine learning operations (MLOps) vendor. That...
    Read article
    Two ships identified by a machine learning model, one of them incorrectly labeled as a sailboat.

    Why, When, and How to Retrain Machine Learning Models

    Once you train a machine learning model effectively, it has the ability to perform incredibly useful tasks, like detecting objects on real-time...
    Read article
    Web of circles and squares illustrating a data lineage process

    What Is Data Lineage?

    Data lineage refers to the full history of data points and actions taken on them throughout a machine learning (ML) workflow. By inspecting a...
    Read article
    Two people on an escalator with their reflections, bounding boxes around them.

    Understanding Performance Bias With the Valor Model Evaluation Service

    Machine learning benchmarks like ImageNet, COCO, and LLM Leaderboard usually target a single metric, such as accuracy for classification tasks or...
    Read article
    Two professionals standing at a dry-erase board.

    To Build or Buy an MLOps Platform?

    It’s the oldest debate in business (or at least the oldest since Silicon Valley invented software-as-a-service). “If we need software, should we buy...
    Read article
    A graphic showing multiple text boxes and a magnifying lens to indicate text classification.

    Text Classification With LLMs: A Roundup of the Best Methods

    This blog post is part of a series on using large language models for text classification. View the other posts here: LLMs for Text Classification:...
    Read article
    Unsupervised text classification

    Unsupervised Text Classification: Categorize Natural Language With LLMs

    This blog post is part of a series on using large language models for text classification. View the other posts here: LLMs for Text Classification:...
    Read article
    Text classification using the Llama model

    LLMs for Text Classification: A Guide to Supervised Learning

    This blog post is part of a series on using large language models for text classification. View the other posts here: Unsupervised Text...
    Read article
    Striveworks secures patent for innovative data lineage process

    Striveworks Secures Patent for Innovative Data Lineage Process

    The new process grants greater transparency into machine learning workflows and allows for better data governance. Originally published on January 4,...
    Read article
    Header image stating

    Navigating AI’s ‘Messy Middle’ in 2024

    With all due apologies to our future president, Taylor Swift, 2023 was the year of AI. The excitement around generative AI crescendoed, seeing vastly...
    Read article
    Overhead photography of cars in a parking lot; a yellow car has been selected for computer vision.

    Demystifying CV: The Power of Convolution in Neural Networks

    Computer vision is an innovative field of research that aims to provide computers with an understanding of digital imagery, often through artificial...
    Read article
    non-separable graph

    Understanding Neural Networks and the Training Process

    Training a neural network involves a lot of mathematics, including linear algebra and multivariate calculus, and a lot of computation. The purpose of...
    Read article
    Two classifiers being combined into a single classifier. In general, the decision boundaries of the two classifiers cannot merely be superimposed; instead new boundaries must be learned that respect both tasks.

    Sequential Learning through Knowledge Distillation

    In the last blog post, we saw that neural networks can be simplified (or pruned) to alleviate their complexity and cost for computation. Often small...
    Read article
    Vehicle color full run model size

    Pruning ResNets for Fun and Profit

    What are Neural Networks? Neural networks are highly parameterized functions which can be trained to fit/represent/model data. Parameterized models...
    Read article