Louis Research group

Interdisciplinary research group |Theoretical Physics at Oxford University| Evolution| Machine learning| Self-assembly| Algorithmic information theory|

[Louis Research group @louis_research]

https://t.co/78ZnVVeqH4

Our pessimistic opinion on the prospect for current QNNs to achieve quantum advantage on classical data.

We also comment on unhelpful characteristics of the QNN literature that need fixing.

[Louis Research group @louis_research]

A fun paper by Yoonsoo Nam, Nayara Fonseca and Seok Hyeong Lee,
https://t.co/RfKVOrSBEa

The model does a good job predicting emergence in a NN on the model task of multi-task sparse-parity

[Petar Veličković @PetarV_93]

After 3 years, it's time for us to start sharing the chapters of the GDL book! ❤️

Also included: companion slides from our @Cambridge_Uni & @UniofOxford courses 🧑‍🎓

Chapter 1 is out **now**!
More to follow soon 🎉

https://t.co/g7SqyZBCgX 📖

@mmbronstein @joanbruna @TacoCohen

[Cees Dekker @cees_dekker]

I have learned biophysics and biology through endlessly asking 'naive' questions.

You'll never learn if you don't ask.

So just ask those naive questions!

[Arlin @ArlinStoltzfus]

The result is that, due to the clumpiness of GP maps, mutational accessibility can matter *even more* than in a simple model of average mut rates (i.e., more than Yampolsky-Stoltzfus)

Here the steeper curves assume a simple model; the flatter ones reflect complex GP maps

3/3

[Arlin @ArlinStoltzfus]

But as Martin, et al. point out, given the clumpiness of real GP maps, mean relative accessibility is misleading

For a given starting genotype for P0, the mut accessibility of an alternative is often 0

This results in overdispersion in arrival rates

2/3

[Arlin @ArlinStoltzfus]

I didn't see the significance of this until @Nora_S_Martin pointed it out just now

Consider adaptive evo from phenotype P0 to one of two options P_r (red) or P_f (blue)

Relative mut rates to red and blue matter, sometimes as much as fitness diffs

1/3
https://t.co/lwzSw6Jrel

[Arlin @ArlinStoltzfus]

Re-upping this for those who didn't see the original

[Chico Camargo @evoluchico]

Have you ever noticed how nature seems to love symmetry?🔺🟥🔴

Evolution has literally trillions of shapes to pick from, and yet, biological structures often show symmetry and simplicity.

This is the story of the discovery that completely changed how I see biology. 🧵

[Louis Research group @louis_research]

https://t.co/QKF3kGfTlO

[Petr Šulc @petr_sulc]

Our new model for DNA and RNA coarse-grained simulations is now out in @JChemPhys :
https://t.co/FapnoqG8ES
It is called #oxNA and includes @ox_DNA and #oxRNA models, and adds support for hybrid DNA:RNA duplexes:
🧵1/4