HNHacker News
TopNewBestAskShowJobs

charleshmartin

83 karma · joined July 5, 2015

Founder, Calculation Consulting Machine Learning and Data Science Expertise

http://www.linkedin.com/in/charlesmartin14

http://calculationconsulting.com

submissionscomments

Detecting Overfit Layers without any Data

twitter.com·1 pts·charleshmartin·
1

Setol: SemiEmpirical Theory of (Deep) Learning

arxiv.org·6 pts·charleshmartin·
3

Setol: A SemiEmpirical of (Deep) Learning

overleaf.com·4 pts·charleshmartin·
1

The Magic of Mistral, a Story of Dragon Kings

calculatedcontent.com·1 pts·charleshmartin·
0

Evaluating Fine-Tuned LLMs with WeightWatcher

calculatedcontent.com·2 pts·charleshmartin·
0

Deep Learning and Effective Correlation Spaces

calculatedcontent.com·1 pts·charleshmartin·
0

Weightwatcher: Data-Free Diagnostics for Deep Learning

weightwatcher.ai·2 pts·charleshmartin·
1

Better than BERT: Pick your best model

calculatedcontent.com·1 pts·charleshmartin·
0

Predicting trends in the quality of state-of-the-art neural networks

nature.com·1 pts·charleshmartin·
0

How to tell if you have trained your model with enough data

calculatedcontent.com·2 pts·charleshmartin·
2

Predicting trends in the quality of state-of-the-art neural networks

nature.com·1 pts·charleshmartin·
0

Is Your Model Overtrained?

calculatedcontent.com·2 pts·charleshmartin·
1

WeightWatcher: Empirical Quality Metrics for Deep Neural Networks

calculatedcontent.com·2 pts·charleshmartin·
0

Towards a New Theory of Learning: Statistical Mechanics of Deep Neural Networks

calculatedcontent.com·5 pts·charleshmartin·
0