# Learning what to remember

arXiv: https://arxiv.org/abs/2201.03806 · HTML: https://mlsanity.com/paper/2201.03806/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 23 references that are in the map.
- First posted on arXiv 11 Jan 2022.

## Builds on (most cited first, 9 of 9)

- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Neural Turing Machines](https://mlsanity.com/paper/1410.5401/) (2014)
- [Memory Networks](https://mlsanity.com/paper/1410.3916/) (2014)
- [Fast Learning Requires Good Memory: A Time-Space Lower Bound for Parity Learning](https://mlsanity.com/paper/1602.05161/) (2016)
- [Not All Memories are Created Equal: Learning to Forget by Expiring](https://mlsanity.com/paper/2105.06548/) (2021)
- [Extractor-Based Time-Space Lower Bounds for Learning](https://mlsanity.com/paper/1708.02639/) (2017)
- [Space lower bounds for linear prediction in the streaming model](https://mlsanity.com/paper/1902.03498/) (2019)
- [Prediction with a Short Memory](https://mlsanity.com/paper/1612.02526/) (2016)
- [Towards a combinatorial characterization of bounded memory learning](https://mlsanity.com/paper/2002.03123/) (2020)

## Abstract (arXiv)

We consider a lifelong learning scenario in which a learner faces a neverending and arbitrary stream of facts and has to decide which ones to retain in its limited memory. We introduce a mathematical model based on the online learning framework, in which the learner measures itself against a collection of experts that are also memory-constrained and that reflect different policies for what to remember. Interspersed with the stream of facts are occasional questions, and on each of these the learner incurs a loss if it has not remembered the corresponding fact. Its goal is to do almost as well as the best expert in hindsight, while using roughly the same amount of memory. We identify difficulties with using the multiplicative weights update algorithm in this memory-constrained scenario, and design an alternative scheme whose regret guarantees are close to the best possible.