Posts in: Blog

JUN 04, 2019

The cloud giants have an AI problem

By Evan Sparks

The general perception of cloud computing is that it makes all compute tasks cheaper and easier to manage.


MAY 20, 2019

Stop doing iterative model development

By Yoav Zimmerman

Imagine a world in which gradient descent or second-order methods have not yet been invented, and the only way to train machine learning models is to tune their weights by hand.


MAR 05, 2019

Random Search is a hard baseline to beat for Neural Architecture Search

By Ameet Talwalkar

In a previous post on “What’s the deal with Neural Architecture Search?”, Liam Li and I discussed Neural Architecture Search (NAS) as a promising research direction that has the potential to replace expert-designed networks with learned, task-specific architectures.


FEB 20, 2019

Addressing the challenges of massively parallel hyperparameter optimization

By Ameet Talwalkar, Desmond Chan

As most deep learning engineers know, it can take days or weeks to train a deep learning model, costing organizations considerable time and money. But what if we could speed up the process and achieve better results in the process?


DEC 18, 2018

What’s the deal with Neural Architecture Search?

By Liam Li, Ameet Talwalkar

Deep learning offers the promise of bypassing the process of manual feature engineering by learning representations in conjunction with statistical models in an end-to-end fashion.


OCT 16, 2018

Warm starting for efficient deep learning resource utilization

By Jennifer Villa, Yoav Zimmerman

In this post, we discuss how warm-starting can save computational resources and improve generalizability when training deep learning models.


AUG 30, 2018

Recapping our first Women in Infrastructure meetup

By Jennifer Villa

Last week, we at Determined AI were honored to sponsor a meetup of the Women in Infrastructure group focused on ML infrastructure.


AUG 20, 2018

Getting the most out of your GPU cluster for deep learning: part II

By Jennifer Villa, Neil Conway

In this post, we discuss the missing key to fully leveraging your hardware investment: specialized software that understands the unique properties of deep learning workloads.


JUL 25, 2018

Getting the most out of your GPU cluster for deep learning: part I

By Jennifer Villa, Jonathan Ben-tzur

To maximize the value of your deep learning hardware, you’ll need to invest in software infrastructure. Setting up a cluster manager is an essential first step in this process, but it’s not the end of the story.


MAY 25, 2018

Reproducibility in ML: why it matters and how to achieve it

By Jennifer Villa, Yoav Zimmerman

Reproducible machine learning is hard, particularly when training deep learning models. We review common sources of DL non-determinism and how to address them.