Related papers: Memory erasure with finite-sized spin reservoir
Dynamic Random Access Memory (DRAM) is the prevalent memory technology used to build main memory systems of almost all computers. A fundamental shortcoming of DRAM is the need to refresh memory cells to keep stored data intact. DRAM refresh…
We study the inherent space requirements of shared storage algorithms in asynchronous fault-prone systems. Previous works use codes to achieve a better storage cost than the well-known replication approach. However, a closer look reveals…
When reformulated as a resource theory, thermodynamics can analyze system behaviors in the single-shot regime. In this, the work required to implement state transitions is bounded by {\alpha}-Renyi divergences and so differs in identifying…
Catastrophic forgetting can be trivially alleviated by keeping all data from previous tasks in memory. Therefore, minimizing the memory footprint while maximizing the amount of relevant information is crucial to the challenge of continual…
The first law of thermodynamics, which governs energy conservation, is traditionally formulated as an equality. Surprisingly, we demonstrate that the first law alone implies a universal Landauer-like inequality linking changes in system…
Data storage systems are more reliable than their individual components. In order to build highly reliable systems out of less reliable parts, systems introduce redundancy. In replicated systems, objects are simply copied several times with…
A quantum message is encoded into $N$ storage nodes (quantum systems $Q_1\dots Q_N$) with assistance from $N_B$ maximally entangled bi-partite quantum systems $A_1B_1, \dots, A_{N_B}B_{N_B}$, that are prepared in advance such that $B_1\dots…
We observe that the yield strength of a variety of materials, including highly structured and densely packed metals, alloys and semi-crystalline polymers is reasonably approximated by the thermal energy density of the material. This…
Landauer's principle imposes a fundamental limit on the energy cost to perfectly initialize a classical bit, which is only reached under the ideal operation with infinite-long time. The question on the cost in the practical operation for a…
The memory effect in a single crystal spin glass ($\mathrm{Cu}_{0.92}\mathrm{Mn}_{0.08}$) has been measured using \freq ac susceptibility techniques over a temperature range of $0.4 - 0.7 \, T_g$ and a model of the memory effect has been…
Conventional wisdom indicates that initial memory should decay away exponentially in time for general (noncritial) equilibration processes. In particular, time-integrated quantities such as heat are presumed to lose initial memory in a…
The energy cost of erasing quantum states depends on our knowledge of the states. We show that learning algorithms can acquire such knowledge to erase many copies of an unknown state at the optimal energy cost. This is proved by showing…
The rapid scaling of artificial intelligence models has revealed a fundamental tension between model capacity (storage) and inference efficiency (computation). While classical information theory focuses on transmission and storage limits,…
One of the major resource requirements of computers - ranging from biological cells to human brains to high-performance (engineered) computers - is the energy used to run them. Those costs of performing a computation have long been a focus…
We examine non-Markovian effects in an open quantum system from the point of view of information flow. To this end, we consider the spin-boson model with a cold reservoir, accounting for the exact time-dependent correlations between the…
The construction of models from data is a significant contributor to the energetic costs of computation. Because of this, understanding how foundational thermodynamic bounds apply to modeling algorithms will be increasingly important. Here,…
Thermodynamics connects our knowledge of the world to our capability to manipulate and thus to control it. This crucial role of control is exemplified by the third law of thermodynamics, Nernst's unattainability principle, which states that…
We propose experimental methods to engineer reservoirs at arbitrary temperature which are feasible with current technology. Our results generalize to mixed states the possibility of quantum state engineering through controlled decoherence.…
In Szilard's engine, a demon measures a one-particle gas and applies feedback to extract work from thermal fluctuations, embodying Maxwell's notion that information reduces thermodynamic entropy - an apparent second-law violation. The…
Scientific computing workflows generate enormous distributed data that is short-lived, yet critical for job completion time. This class of data is called intermediate data. A common way to achieve high data availability is to replicate…