The Simply Uninformed Thermodynamics of Erasure
John D. Norton

TL;DR
This paper clarifies the thermodynamic costs of different erasure types, showing that strong erasure inevitably produces entropy due to thermal noise suppression, while weak erasure can be dissipationless.
Contribution
It distinguishes strong and weak erasure notions and analyzes their thermodynamic costs, correcting misconceptions about entropy in the Gibbs formalism.
Findings
Strong erasure cannot be dissipationless.
Entropy creation is mainly due to thermal noise suppression.
Weak erasure can have zero minimum entropy cost.
Abstract
1. Strong and weak notions of erasure are distinguished according to whether the single erasure procedure does or does not leave the environment in the same state independently of the pre-erasure state. 2. Purely thermodynamic considerations show that strong erasure cannot be dissipationless. 3. The main source of entropy creation in erasure processes at molecular scales is the entropy that must be created to suppress thermal fluctuations ("noise"). 4. A phase space analysis recovers no minimum entropy cost for weak erasure and a positive minimum entropy cost for strong erasure. 5. An information entropy term has been attributed mistakenly to pre-erasure states in the Gibbs formalism through the neglect of an additive constant in the "-k sum p log p" Gibbs entropy formula.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
