tweetindex
IT

Jorge Bravo Abad ✓

@bravo_abad · joined 20 Oct 2013

Building closed loops that turn scientific discovery into infrastructure · Director, AI for Materials Lab @UAM_Madrid · Three books on AI & science (1 textbook)

12 105Followers
10 188Following
3 682Posts total
4.4KViews on collected posts

Ultimi post

In AI for Science, "ground truth" is rarely ground truth. Machine learning language makes labels sound definitive: inputs go in, the "correct answer" comes out, and the model is judged by how closely it matches it. But in science, that answer is usually itself a measurement, an
452 views · 12 likes · 0 reposts · 3 replies Open on X →
The prior can matter more than the downstream model: a lesson from gene regulatory networks When reconstructing a biological network, we usually focus on the model that analyzes the experimental data. This paper shows that an earlier step can matter just as much: deciding which
607 views · 13 likes · 3 reposts · 0 replies Open on X →
Your test split defines your scientific claim. A high test score can answer the wrong scientific question. Imagine your dataset contains repeated measurements from the same sample, molecules from the same chemical family, nearby time points from the same trajectory, or https://
842 views · 21 likes · 1 reposts · 2 replies Open on X →
@bravo_abad Worth pausing on lower as the success metric. A search that samples more geometries will find lower barriers simply because it looked harder. The 61 to 80 move on fidelity is what shows the paths hold up. 8 views · 0 likes · 0 reposts · 0 replies Open on X →
Scaling chemical intuition: automating reaction-path discovery in catalysis Machine-learned interatomic potentials have made atomistic calculations dramatically faster. But that does not automatically make the scientific workflow autonomous. For surface reactions, someone still
952 views · 17 likes · 4 reposts · 1 replies Open on X →
@bravo_abad Por fin un libro que convierte a los científicos en dueños de la IA y no en simples usuarios. 14 views · 0 likes · 0 reposts · 0 replies Open on X →
Hoy es el día: ya está disponible "Inteligencia artificial para estudiantes de ciencias" (Ediciones Pirámide, Grupo Anaya). 📘 Es el libro de IA que me habría gustado tener cuando empecé a aplicar aprendizaje automático a problemas científicos. Está pensado para estudiantes e ht
1.5K views · 9 likes · 4 reposts · 1 replies Open on X →

Rispetto ad account della stessa dimensione

7 post degli ultimi 90 giorni, accanto alla fascia di 10K–100K follower. arriva ben oltre il proprio pubblico, e quel pubblico reagisce.

Visualizzazioni mediane607questo account924mediana per 10K–100K
Copertura, %5.01%questo account3.62%mediana per 10K–100K
Interazione, %2.31%questo account1.52%mediana per 10K–100K
MetricaQuesto accountMediana per 10K–100KRapporto
Visualizzazioni mediane per post6079240.66×
Copertura (visualizzazioni ÷ follower)5.01%3.62%1.39×
Tasso di interazione2.31%1.52%1.52×

Altri account di questa fascia →   Confronta con un altro account →   Come sono costruiti questi parametri →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

1.5K3 Sep
14
95214 Sep
8
842
60715 Sep
452

Last 7 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.95%3 Sep
0.00%
2.31%14 Sep
0.00%
2.97%
2.64%15 Sep
3.32%

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.52%.

What the audience does

Likes61.0%72 in total
Reposts10.2%12 in total
Replies5.9%7 in total
Quotes0.8%1 in total
Bookmarks22.0%26 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Account simili