←
Brett David Roads13 posts
Brett David Roads avatar

Brett David Roads

@BDRoads

he/him, Terran, Postdoctoral Research Associate

◷ Joined January 2019
George Monbiot avatar
⟳ Brett David Roads retweeted
George Monbiot@GeorgeMonbiot·20 Sep 2022

I tried to explain it in 2014. Despite its many and massive failures since then, it is now more powerful than ever before. theguardian.com/books/2016/a…

Bradley Love avatar
⟳ Brett David Roads retweeted
Bradley Love@ProfData·18 Feb 2021

The "The Costs and Benefits of Goal-Directed Attention in Deep Convolutional Neural Networks" is now out in Computational Brain and Behavior. The quote tweet below briefly walks through the preprint. The published version includes an added bonus, link.springer.com/article/10… (1/2)

Post attachment

People use top-down attention, such as when searching for one's keys. One idea is that prefrontal cortex helps reconfigure our visual system to be more sensitive to features relevant to our goals. This has costs and benefits. 2/n

Bradley Love avatar
⟳ Brett David Roads retweeted
Bradley Love@ProfData·15 Feb 2021

New preprint, "A Too-Good-to-be-True Prior to Reduce Shortcut Reliance". If it's too good to be true, it probably is and that holds for deep learning as well. To generalize broadly, models need to learn invariants but instead are fooled by shortcuts. arxiv.org/abs/2102.06406 (1/4)

Post attachment
Bradley Love avatar
⟳ Brett David Roads retweeted
Bradley Love@ProfData·24 Nov 2020

Introducing the embedding space you didn't know you needed: Human similarity judgments for the entire ImageNet (50k images) validation set. Perfect for evaluating representations, including unsupervised models. It's already bearing fruit, w @BDRoads arxiv.org/abs/2011.11015 (1/3)

Post attachment
Bradley Love avatar
⟳ Brett David Roads retweeted
Bradley Love@ProfData·23 Oct 2020

New blog, "A neuroscience-inspired approach to transfer learning" w @ken_lxl @BDRoads. We add goal-directed attention to the middle of a deep convolutional network and find it better adapts to new tasks than retraining the top layer as is standard in ML. bradlove.org/blog/attention

Bradley Love avatar
⟳ Brett David Roads retweeted
Bradley Love@ProfData·17 Jan 2020

New paper w @BDRoads, "Learning as the Unsupervised Alignment of Conceptual Systems". Supervised learning tasks can be solved by purely unsupervised means by exploiting correspondences across systems (e.g., text, images, etc.). 1/5 nature.com/articles/s42256-0…

Post attachment