The flip side of very fast progress is that it's typically happening due to a phase transition, and it will inevitably be followed by a period of diminishing returns (until the next phase transition)
@fchollet
-
Mini-challenge: Win signed Deep Learning with Python 2E book
By
–
Mini-challenge: If you score > 0.3 on the leaderboard with a fork of this notebook, I'll send you a signed copy of Deep Learning with Python 2E
-
TFData Pipeline Performance and Functional Composition Showcase
By
–
This sort of unusual dataset makes a great showcase for the power of TFData 🙂 All-functional chaining, filters operations, super readable, etc. And you get something ultra-performant that compiles to parallel C++ (no Python at all at runtime) Here's my randomized pipeline…
-
End-to-end ML pipeline with Keras U-Net and TFData implementation
By
–
The model is pretty weak right now, but the end-to-end pipeline works. It trains on the whole dataset (a bit of a challenge given the memory constrained and the fact that streaming isn't an option!) and makes a submission. It uses TFData for the pipeline and a Keras U-Net.
-
New Kaggle Competition: Reading Carbonized Papyrii from Pompeii
By
–
I just wrote up a quick notebook for the new Kaggle competition on "reading" carbonized papyrii from Pompeii via x-ray 3D scans. Really fun topic 🙂
-
Parameters Storage: Understanding Models Beyond Traditional Databases
By
–
Only the parameters. The data isn't stored (though that distinction isn't that important). In that sense it's more like a regression curve than a dict. It is not an exact superset of a RDBMS.
-
KerasNLP Reaches 30,000 Monthly Downloads Milestone
By
–
KerasNLP adoption is still early days, but doing ok! We just crossed 30,000 downloads / month. https://
keras.io/api/keras_nlp/ -
Neural Networks as Interpolative Databases with Natural Language Interfaces
By
–
You can retrieve not just what was seen at training time, but arbitrary combinations of it. It's an interpolative database and program store, with a natural language interface.
-
LLMs as Queryable Databases, Not Mere Autocomplete Systems
By
–
"It's autocomplete" is not a helpful analogy to understand LLMs. A LLM is more like a database that lets query information in natural language. You can query both knowledge, and "patterns" (associative programs seen in the training data, that can be applied to new inputs).
-
Replication and AGI: Equal Distance from Current Reality
By
–
We are as far today from arbitrary matter replication and grey goo as from AGI.