vackosar

6 ms·

1.
Bellman Update and Synthetic Data in Q-Transformer(vaclavkosar.com)
1 points by vackosar 3y ago|0 comments
2.
Synthetic Data for LLM Training(vaclavkosar.com)
1 points by vackosar 3y ago|0 comments
3.
by vackosar 3y ago
Egan is great. Another good book is Incandescence. Only problem is that some of his novels contain some kind of loneliness. Can't describe it.
4.
by vackosar 3y ago
I see only the problem of no ability to change the OKRs, otherwise the core idea of making a plan is right. Changing the goals should introduce a additional change process, because priorities change with new knowledge. I haven't see th
5.
How to Install Your Python Version on Ubuntu(vaclavkosar.com)
1 points by vackosar 3y ago|2 comments
6.
Dangers of Python Lambda: Repeated Values Due to Late Binding(vaclavkosar.com)
2 points by vackosar 3y ago|0 comments
7.
by vackosar 3y ago
I use mapy.cz often, because of the UI and better maps. Only the traffic is probably not estimated the best.
8.
Validate Reliability of a Paper: authority, bias, evidence, comparison, impact(vaclavkosar.com)
1 points by vackosar 3y ago|0 comments
9.
by vackosar 3y ago
I read about criticism of General Relativity, that I haven't verified so grain of salt please. The criticism is in that the equations are so complex that only the simplest limits of them are studied and so the theory is therefore not v
10.
by vackosar 3y ago
https://vaclavkosar.com/ Mostly on machine learning research
11.
Prompting Techniques That Squeeze the Best Out of Your LLM(vaclavkosar.com)
2 points by vackosar 3y ago|0 comments
12.
by vackosar 3y ago
True problem is not technical and it is right there in the article. Non-academics "speak less and worry more". So their knowledge is likely scattered and not distilled. Building these devices is probably quite simple in the end, i
13.
Reasoning and reducing hallucinations in LLMs by hybridizing with algorithms(vaclavkosar.com)
1 points by vackosar 3y ago|0 comments
14.
The Bridge Between Sync and Async World(medium.com)
1 points by vackosar 4y ago|0 comments
15.
Hands on Machine Learning with Multilingual Image-Text Fashion Dataset(colab.research.google.com)
1 points by vackosar 4y ago|0 comments
16.
Fill Versions from Python Environment Using This Script(vaclavkosar.com)
1 points by vackosar 4y ago|0 comments
17.
Glami-1M: Multilingual Image-Text Fashion Dataset(medium.com)
2 points by vackosar 4y ago|0 comments
18.
Glami-1M: A Multilingual Image-Text Fashion Dataset(github.com)
1 points by vackosar 4y ago|0 comments
19.
Neural Network Pruning Explained(vaclavkosar.com)
2 points by vackosar 4y ago|0 comments
20.
Multimodal Image-Text Classification: Understand the Best Models(vaclavkosar.com)
1 points by vackosar 4y ago|0 comments
21.
Python Functools.cmp_to_key Explained(vaclavkosar.com)
2 points by vackosar 4y ago|0 comments
22.
Strong Static Typing vs. Weak Dynamic Typing(vaclavkosar.com)
1 points by vackosar 4y ago|0 comments
23.
Neural Data Compression: Reducing bits with deep-learning(vaclavkosar.com)
2 points by vackosar 4y ago|0 comments
24.
OpenAI's DALL-E 2 and DALL-E 1 Explained(vaclavkosar.com)
2 points by vackosar 4y ago|0 comments
25.
by vackosar 4y ago
Correction! Cost is around $10M not $10B.
26.
by vackosar 4y ago
Sounds interesting! Would you link to that or describe them here? Thanks!
27.
by vackosar 4y ago
Yes, of course, thanks!
28.
by vackosar 4y ago
Correction!! The model costed around 10M not 10B! Thanks for raising that. Mistake during copying from the second slide :(
29.
Google's Pathways Language Model and Chain-of-Thought(vaclavkosar.com)
77 points by vackosar 4y ago|30 comments
30.
by vackosar 4y ago
I write on machine learning, software, business, and many other things. https://vaclavkosar.com/

More ›