Law
Predictive coding in balanced neural networks with noise, chaos and delays
However, can one do better?
Remember What You Want to Forget: Algorithms for Machine Unlearning Ayush Sekhari Cornell University
We study the problem of unlearning datapoints from a learnt model.
Algorithms that Approximate Data Removal: New Results and Limitations
The proliferation of techniques that employ user data to do things such as training and validating machine learning across a variety of organizations has led to reconsideration of how to interpret RtbF.
Watermarking Makes Language Models Radioactive Tom Sander
Current methods like membership inference or active IP protection either work only in settings where the suspected text is known or do not provide reliable statistical guarantees. We discover that, on the contrary, it is possible to reliably determine if a language model was trained on synthetic data if that data is output by a watermarked LLM.
A Fair Classifier Using Kernel Density Estimation
It is now employed to make critical decisions that affect our lives, cultures, and rights, e.g., screening job applicants, and informing bail & parole decisions.
