# Judy Hoffman

En popüler podcast bölümleri
Güncellenmiş:

[S2 Re-sliced] 39. What are some traits common across some of the best collaborators you've worked with?

Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited abo…

[S2 Re-sliced] 40. Describe something that made you smile today.

Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited abo…

[S2 Re-sliced] 41. What is some of the best advice you've gotten or given?

Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited abo…

[S2 Re-sliced] 42. Why did you agree to do this interview with me?

Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited abo…

[S2 Re-sliced] 43. Is there anything you'd like to talk about that we haven't covered?

Humans of AI: Stories, Not Stats is an interview series with AI researchers to get to know them better as people. We don't talk about AI or their work or the stats of their life like what college they went to. They share what they think about, what they are insecure about, what they get excited abo…

Token Merging: Your ViT But Faster

We introduce Token Merging (ToMe), a simple method to increase the throughput of existing ViT models without needing to train. ToMe gradually combines similar tokens in a transformer using a general and light-weight matching algorithm that is as fast as pruning while being more accurate. Qualitativ…

ZipIt! Merging Models from Different Tasks without Training

Typical deep visual recognition models are capable of performing the one task they were trained on. In this paper, we tackle the extremely difficult problem of combining completely distinct models with different initializations, each solving a separate task, into one multi-task model without any ad…

Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles

Modern hierarchical vision transformers have added several vision-specific components in the pursuit of supervised classification performance. While these components lead to effective accuracies and attractive FLOP counts, the added complexity actually makes these transformers slower than their van…

arxiv preprint - ZipIt! Merging Models from Different Tasks without Training

In this episode we discuss ZipIt! Merging Models from Different Tasks without Training by George Stoica, Daniel Bolya, Jakob Bjorner, Taylor Hearn, Judy Hoffman. The paper introduces a method called "ZipIt!" that can merge two deep visual recognition models trained on separate tasks without add…

Our Daily Show! Ft: Clay Marquis, Luis Nolla, and More!

Clay Marquis joins us in the 9 to bring us insights on "God's Portion" written by his mother Judy Hoffman. Clay will take us through the all of the work it took for Judy to formulate the book. Plus Clay will give us insights into his relationship to his Step-Father Lars Hoffman, and the beautiful r…