Gary Marcus on Twitter: "= yet another a promissory note. "DNNs represent a powerful framework for building task-performing models and will drive substantial insights in computational neuroscience." NOT: "have driven" Maybe it will pan out, maybe not. It's your certainty that I keep objecting to.… https://t.co/X0llbaBR1j"

Country	Code	For customers of
United States	40404	(any)
Canada	21212	(any)
United Kingdom	86444	Vodafone, Orange, 3, O2
Brazil	40404	Nextel, TIM
Haiti	40404	Digicel, Voila
Ireland	51210	Vodafone, O2
India	53000	Bharti Airtel, Videocon, Reliance
Indonesia	89887	AXIS, 3, Telkomsel, Indosat, XL Axiata
Italy	4880804	Wind
3424486444	Vodafone
» See SMS short codes for other countries

ricardo chavarriaga‏ @r_chavarriaga Nov 29

Report Tweet

Replying to @r_chavarriaga @KordingLab and

I'm a big believer that #ANNs are a great tool to improve our knowledge of the brain, but often think the "ANN as abstraction of the brain" analogy does more harm than benefit /end

2 replies 2 retweets 7 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @r_chavarriaga @KordingLab and

You need to separate your concern about hype in the popular press from the use of ANNs in neuroscience. ANNs are a very useful model in neuroscience, and we shouldn't shy away from that because Deep Mind's PR ppl go a bit far sometimes.

2 replies 0 retweets 7 likes

Gary Marcus‏ @GaryMarcus Nov 29

Report Tweet

Replying to @tyrell_turing @r_chavarriaga and

what have ANNs done for neuroscience, aside from serving as classifiers?

3 replies 0 retweets 1 like

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @GaryMarcus @r_chavarriaga and

You don't actually keep up with the neuroscience literature, do you? That has been evident in these conversations... Here, lemme give you a few examples:

2 replies 2 retweets 34 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

1) ANNs optimised on relevant tasks match the representations in human (and primate) cortical areas better than other models developed to date: https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003963 … https://www.sciencedirect.com/science/article/pii/S0896627318302502 …https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1003915 …

1 reply 7 retweets 45 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

2) ANNs trained on motor tasks successfully predict both motor behaviour and the distribution of representations in motor cortex: https://www.sciencedirect.com/science/article/pii/S0896627312009920 … https://www.biorxiv.org/content/biorxiv/early/2019/08/24/742189.full.pdf …

1 reply 0 retweets 27 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

3) Recurrent ANNs can model prefrontal cortical activity well, and explain both behaviour and representations related to multi-task learning: https://www.nature.com/articles/s41593-018-0310-2 …https://www.nature.com/articles/s41593-018-0147-8 …

2 replies 0 retweets 19 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

4) In my own work, I found that the original predictions of complimentary learning theory from ANNs (https://psycnet.apa.org/record/1995-42327-001 …) predicts the impact of memory consolidation on mouse navigation behaviour:https://www.nature.com/articles/nn.3736 …

3 replies 0 retweets 19 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

And there's more! But, I have to go to lab meeting... Nonetheless, I would encourage you to actually approach the role of ANNs in neuroscience with an open mind. A lot has been accomplished, and there's more to come.

1 reply 1 retweet 14 likes

Blake Richards‏ @tyrell_turing Nov 29

Report Tweet

Replying to @tyrell_turing @GaryMarcus and

Here's a couple of reviews for you if you're actually interested and not so ideological that you are incapable of approaching it neutrally (you have not convinced me of this to date): https://www.nature.com/articles/s41593-019-0520-2 …https://www.biorxiv.org/content/10.1101/133504v2.abstract …

4 replies 0 retweets 12 likes

Gary Marcus‏ @GaryMarcus Nov 29

Report Tweet

Replying to @tyrell_turing @r_chavarriaga and

= yet another a promissory note. "DNNs represent a powerful framework for building task-performing models and will drive substantial insights in computational neuroscience." NOT: "have driven" Maybe it will pan out, maybe not. It's your certainty that I keep objecting to.

9:16 AM - 29 Nov 2019

4 Likes

1 reply 0 retweets 4 likes

1. Brad Wyble‏ @bradpwyble 23 hours ago
  
  Report Tweet
  Replying to @GaryMarcus @tyrell_turing and
  
  For me, ANNs have provided an intuition for how at least some aspects of object recognition might be solved. It's far from complete but it's also more than the promissory note that early workmon perceptrons provided
  
  7 likes
  Thanks. Twitter will use this to make your timeline better. Undo
  
  Undo

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

Tweets

Gary Marcus

@GaryMarcus

Tweets

Loading seems to be taking a while.

false

Saved searches

Tweets

Gary Marcus

@GaryMarcus

Tweets

Go to a person's profile

Saved searches

Promote this Tweet

Block

Tweet with a location

Your lists

Create a new list

Copy link to Tweet

Embed this Tweet

Embed this Video

Preview

Why you're seeing this ad

Log in to Twitter

Sign up for Twitter

Not on Twitter? Sign up, tune into the things you care about, and get updates as they happen.

Two-way (sending and receiving) short codes:

Confirmation

Welcome home!

Tweets not working for you?

Say a lot with a little

Spread the word

Join the conversation

Learn the latest

Get more of what you love

Find what's happening

Never miss a Moment

Loading seems to be taking a while.

Promoted Tweet

false