Vy Vo
@vvobot
Followers
604
Following
431
Media
14
Statuses
339
Brain-Inspired AI research @ Intel Labs. Neurosciences PhD.
Joined May 2014
Happy to share that our empirical work co-led by Shailee Jain on a method to use deep learning models for neuroscience inferences (specifically, to understand temporal context effects) is available as a preprint!.
🌟NEW PREPRINT🌟.🤖-based🧠models are all the rage!. But,.- How do we interpret 🧠if 🤖are SO un-interpretable?.- How do we _test_ if🤖are good models of 🧠? (No, your prediction score is NOT the answer). Our new paper (co-lead @vvobot) shows you how!. 1/9.
0
0
12
RT @sucholutsky: Announcing the First Representational Alignment Workshop!. RSVP - Organized by: @sucholutsky @dre….
0
46
0
Very cool finding!.
Excited to share our new preprint with @mtoneva1, @ptoncompmemlab, and @manojneuro), in which we ask if GPT-3 (a large language model) can segment narratives into meaningful events similarly to humans. We use an unconventional approach: ⬇️.
0
0
4
Looking forward to our workshop tomorrow at NeurIPS! #intelai.
Come over to the MemARI workshop at #NeurIPS2022 tomorrow Dec 2 to chat about how memory mechanisms should be designed in deep learning and what this field can learn from biological memory systems. Schedule: Website:
0
0
4
Looking forward to discussing our NeurIPS workshop paper and more in person on Friday!.
Transformers are limited in their ability to generalize over long temporal dependencies. In our @NeurIPSConf paper (#MemARI), we propose a framework for effectively integrating findings from human memory into LMs / transformers. (1/2).
0
1
4
RT @shaileeejain: **FINAL** deadline extension for the #MemARI workshop @ #NeurIPS2022!. Thurs | Sept 29th, 2022 | 23:59 hrs AOE.
0
4
0
RT @shaileeejain: Meet the #MemARI speakers: Hava Siegelmann, @HochreiterSepp, @kanileke, @criticalneuro & Sainbayar Sukhbaatar! Our speake….
0
5
0
RT @shaileeejain: Interested in memory in 🧠&🤖? Often think about how to build better mem mechanisms in 🤖 or how the 🧠 can help? Our new #ne….
0
18
0
RT @TheCharleyWu: 🚨Preprint alert! An influential theory by @AlisonGopnik describes human development to stochastic….
0
75
0
RT @ptoncompmemlab: Now out in @ApertureOHBM! Our comprehensive overview, led by @manojneuro, of the @UseBrainIAK toolkit for advanced fMRI….
0
18
0
RT @neuralreckoning: If you're interested in the sort of research that I'm interested in (e.g. computational/theory work in SNNs or the aud….
0
10
0
We're looking for a summer 2022 research intern for a AI/cog/neuro effort! If you are interested in some combination of language, context, and memory, send me a message. We have different project ideas for different backgrounds!.
9
48
185
RT @TimKietzmann: Tweeps, I am hiring (big time) - please spread the word. I am looking for three postdocs in the space of neuro-inspired….
0
247
0
The official poster session is over now, but please feel free to reach out here or over email with any questions or comments, or if you want to have a discussion. Happy to have brief chats or longer meetings!.
0
0
1
Summarized in more detail here by equal contributor @shaileeejain for our #SNL2021 presentation: And many thanks to the rest of the team -- Nicole Beckage, @sherryhyc, Chiadika Obinwa & @alex_ander.
2/ In our NeurIPS’20 paper, we showed how timescale-controlled LSTMs can estimate the processing timescale across the 🧠 But can this model implicitly capture popular (and clever) fMRI-studies to investigate timescales? (Specifically from @HassonLab & @chrishoney's group)
1
2
5
We used a multi-timescale LSTM language model to predict fMRI responses while subjects listened to stories. Using in-silico experiments with this predictive model, we show that we can largely capture previously reported effects of temporal context / timescale.
1
0
6
RT @SteinmetzNeuro: The Steinmetz Lab at UW Seattle and the International Brain Laboratory (@IntlBrainLab) seek a research scientist to joi….
0
69
0
RT @postlelab: Preview: Vol.34 Iss. 1 of the Journal of Cognitive Neuroscience will include this assessment of first year of the gender cit….
0
9
0