Neural Network Pruning and Training with Jonathan Frankle at MosaicML

Neural Network Pruning and Training with Jonathan Frankle at MosaicML

Author: Lukas Biewald April 4, 2023 Duration: 1:02:00

Jonathan Frankle, Chief Scientist at MosaicML and Assistant Professor of Computer Science at Harvard University, joins us on this episode. With comprehensive infrastructure and software tools, MosaicML aims to help businesses train complex machine-learning models using their own proprietary data.

We discuss:

- Details of Jonathan’s Ph.D. dissertation which explores his “Lottery Ticket Hypothesis.”

- The role of neural network pruning and how it impacts the performance of ML models.

- Why transformers will be the go-to way to train NLP models for the foreseeable future.

- Why the process of speeding up neural net learning is both scientific and artisanal.

- What MosaicML does, and how it approaches working with clients.

- The challenges for developing AGI.

- Details around ML training policy and ethics.

- Why data brings the magic to customized ML models.

- The many use cases for companies looking to build customized AI models.

Jonathan Frankle - https://www.linkedin.com/in/jfrankle/

Resources:

- https://mosaicml.com/

- The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

Thanks for listening to the Gradient Dissent podcast, brought to you by Weights & Biases. If you enjoyed this episode, please leave a review to help get the word out about the show. And be sure to subscribe so you never miss another insightful conversation.

#OCR #DeepLearning #AI #Modeling #ML


Lukas Biewald hosts Gradient Dissent: Conversations on AI, a series that moves beyond theoretical discussions to examine how artificial intelligence is actually built and deployed. Each episode features a direct, unscripted talk with a leading practitioner-you’ll hear from engineers and researchers at places like NVIDIA, Meta, Google, Lyft, and OpenAI. The focus is on the tangible challenges and breakthroughs they encounter, from initial research to the complex reality of putting models into production. This isn't about abstract futures; it's a grounded look at the decisions shaping the field right now. Biewald, bringing his perspective from Weights & Biases, steers conversations toward the practical trade-offs and collaborative efforts that define modern AI work. For anyone in technology or business who wants to understand the mechanics behind the headlines, this podcast offers a rare, candid window into the process. You’ll come away with a clearer sense of how ideas become functional systems and what it really takes to operate at the cutting edge.
Author: Language: English Episodes: 100

Gradient Dissent: Conversations on AI
Podcast Episodes
Scaling LLMs and Accelerating Adoption with Aidan Gomez at Cohere [not-audio_url] [/not-audio_url]

Duration: 51:31
On this episode, we’re joined by Aidan Gomez, Co-Founder and CEO at Cohere. Cohere develops and releases a range of innovative AI-powered tools and solutions for a variety of NLP use cases.We discuss:- What “attention” m…
Shreya Shankar — Operationalizing Machine Learning [not-audio_url] [/not-audio_url]

Duration: 54:38
About This EpisodeShreya Shankar is a computer scientist, PhD student in databases at UC Berkeley, and co-author of "Operationalizing Machine Learning: An Interview Study", an ethnographic interview study with 18 machine…
Jeremy Howard — The Simple but Profound Insight Behind Diffusion [not-audio_url] [/not-audio_url]

Duration: 1:12:57
Jeremy Howard is a co-founder of fast.ai, the non-profit research group behind the popular massive open online course "Practical Deep Learning for Coders", and the open source deep learning library "fastai".Jeremy is als…
Jerome Pesenti — Large Language Models, PyTorch, and Meta [not-audio_url] [/not-audio_url]

Duration: 52:35
Jerome Pesenti is the former VP of AI at Meta, a tech conglomerate that includes Facebook, WhatsApp, and Instagram, and one of the most exciting places where AI research is happening today.Jerome shares his thoughts on T…
D. Sculley — Technical Debt, Trade-offs, and Kaggle [not-audio_url] [/not-audio_url]

Duration: 1:00:26
D. Sculley is CEO of Kaggle, the beloved and well-known data science and machine learning community.D. discusses his influential 2015 paper "Machine Learning: The High Interest Credit Card of Technical Debt" and what the…