Partial annealing and pattern decorrelation in associative neural networks Publication type: Paper
Linda Albanese 1,2,∗ \orcid 0000-0003-1783-0775, Andrea Alessandrelli 1,2 \orcid 0000-0003-4583-1417, Adriano Barra 3,4,5 \orcid 0000-0003-4255-7678, Silvio Franz 1,2,6 \orcid 0000-0001-8300-8443 and Federico Ricci-Tersenghi 4,7,8 \orcid 0000-0003-4970-7376 · 2026
Read Partial annealing and pattern decorrelation in associative neural networks Publication type: Paper by Linda Albanese 1,2,∗ \orcid 0000-0003-1783-0775, Andrea Alessandrelli 1,2 \orcid 0000-0003-4583-1417, Adriano Barra 3,4,5 \orcid 0000-0003-4255-7678, Silvio Franz 1,2,6 \orcid 0000-0001-8300-8443 and Federico Ricci-Tersenghi 4,7,8 \orcid 0000-0003-4970-7376 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-10304.txt.
Introduction
Linda Albanese and four colleagues study memory networks that retrieve stored patterns from partial cues. Using the Hopfield model as a test case, their 2026 paper examines partial annealing. Fast neural dynamics couple to slowly shifting patterns in a two-temperature, two-timescale setup. A real number n tunes the time split and the link between neurons and synapses.
Adapting Guerra's interpolation, the authors derive the free energy without analytic continuation, a delicate analytic step. They show that negative values of n slowly uncorrelate stored patterns, cutting cross-talk and favoring orthogonal layouts. That pushes the net to maximal storage, one pattern per neuron. Mean-field Monte Carlo tests confirm the picture, and partial annealing restores recall even with biased patterns, beating standard uncorrelation methods.
Catalog record
| CREATOR | Linda Albanese 1,2,∗ \orcid 0000-0003-1783-0775, Andrea Alessandrelli 1,2 \orcid 0000-0003-4583-1417, Adriano Barra 3,4,5 \orcid 0000-0003-4255-7678, Silvio Franz 1,2,6 \orcid 0000-0001-8300-8443 and Federico Ricci-Tersenghi 4,7,8 \orcid 0000-0003-4970-7376 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.10304
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Partial annealing and pattern decorrelation in associative neural networks Publication type: Paper Linda Albanese 1,2,∗ \orcid 0000-0003-1783-0775, Andrea Alessandrelli 1,2 \orcid 0000-0003-4583-1417, Adriano Barra 3,4,5 \orcid 0000-0003-4255-7678, Silvio Franz 1,2,6 \orcid 0000-0001-8300-8443 and Federico Ricci-Tersenghi 4,7,8 \orcid 0000-0003-4970-7376 Abstract Using the Hopfield model as a benchmark case, the present work focuses on the investigation of partially annealed associative neural networks, wherein neural dynamics is coupled to slowly evolving patterns within the two-temperature-two-timescale framework. This setting inherently introduces a real parameter n n , reminiscent of the number of replicas in the celebrated replica trick, that tunes the separation of timescales and the effective interaction between fast (i.e. the neurons) and slow (i.e. the synapses) degrees of freedom. By adapting Guerra’s interpolation to the case, we derive the free energy without relying on analytical continuation. The obtained results demonstrate that negative values of n n induce a progressive decorrelation of the stored patterns, thereby effectively reducing interference, promoting orthogonal configurations and ultimately conferring to the network the maximal storage α c = 1 \alpha_{c}=1 . Numerical simulations based on a mean field Monte Carlo dynamics have been employed to confirm this scenario and prove that partial annealing restores retrieval in challenging regimes, such as in the presence of biased patterns, outperforming standard decorrelation methods. These findings underscore the notion of partial annealing as an adaptive mechanism for enhancing memory organisation and retrieval in complex systems. keywords Partial annealing, Neural networks, Guerra interpolation, Statistical mechanics 1 Introduction The statistical mechanics of complex systems has long constituted an effective framework for the quantitative analysis of associative memory networks, in particular for studying pattern recognition, memory storage, and information processing in neural networks [ 32 , 28 , 9 , 30 , 20 ] . Disordered systems characteristically manifest intricate energy landscapes, exhibiting a multitude of metastable configurations and frustration [ 32 , 33 , 25 ] . In statistical mechanics, a fundamental conceptual distinction is made in the description of such systems between quenched and annealed disorder . In the quenched scenario, disorder variables evolve on timescales that are much longer than those associated with the dynamical degrees of freedom. As a result, on the pratical level, they can be regarded as effectively frozen. Conversely, in annealed systems, both sets of variables equilibrate on comparable timescales. While these two limiting cases are analytically tractable, many real systems operate in an intermediate regime, in which disorder evolves slowly but cannot be considered as completely frozen [ 18 , 38 ] . In these regards, despite the theoretical analysis of these systems via the replica method was carried on already a long time ago [ 22 ] , some approaches such as the Guerra interpolation [ 27 , 31 , 11 , 8 , 7 ] , which have played a pivotal role in determining the thermodynamic properties of mean field spin glass models, were lacking for partial annealing and are developed in this paper. Moreover, the theoretical analysis of these systems is founded on a range of analytical techniques, including the replica method and interpolation schemes such as the Guerra interpolation [ 27 , 31 , 11 , 8 , 7 ] , which have played a pivotal role in determining the thermodynamic properties of mean field spin glass models. The concept of partial annealing [ 22 ] was introduced to capture this intermediate regime with precision. In such systems, slow variables, which are often associated with interactions or couplings, are not fixed but evolve according to an effective thermodynamic principle. This is typically characterized by a temperature distinct from that one governing the fast dynamical variables. This approach was originally proposed as a means to study disordered systems in which the couplings themselves are capable of adapting or reassessing, giving rise to effects such as overfrustration and non-trivial feedback between slow and fast degrees of freedom (see [ 22 ] ). The incorporation of two interacting thermal processes within the framework of partial annealing naturally interpolates between the quenched and annealed limits. In the realm of theoretical physics, the statistical mechanics of partially annealed systems can be formulated within the two temperatures framework. In this setting, the fast degrees of freedom (i.e. the neurons) are considered to reach equilibrium at a given temperature, whereas the slow variables (i.e. the synapses) are assumed to evolve at a separate (colder) temperature, hence their dynamics is expected to be slower with respect to the neural one [ 19 , 38 ] . The effective free energy governing the slow variables is thus shown to involve the partition function of the fast subsystem, raised to a power determined by the ratio between the two temperatures. This configuration lends the replica method particular analytical efficacy, as the temperature ratio can be interpreted as an effective replica number, thereby facilitating the application of standard techniques developed for the study of disordered systems. In accordance with the established framework, the analysis is conventionally conducted under the Replica Symmetric (RS) assumption, which entails the treatment of all replicas as equivalent in the saddle point solution. However, in many disordered systems, this symmetry may become unstable [ 21 ] , requiring the more general Replica Symmetry Breaking (RSB) scheme [ 32 , 1 ] , which captures the hierarchical organization of metastable states in the free energy landscape. This formalism was soon applied to the study of adaptive and neural systems, in which synaptic interactions may evolve on timescales that are longer than those characterizing neuronal dynamics. In the early stages of research, analyses were conducted on neural networks with partially annealed couplings. These analyzes demonstrated that the adaptation of interactions has the capacity to cause substantial alterations in both the structure of the attractors and the phase diagram of the system [ 24 ] . Later investigations examined these effects in more specific neural network models, emphasising how the interaction between slow synaptic evolution and fast neural activity can give rise to non-trivial forms of collective behaviour [ 37 ] . In recent years related concepts have subsequently emerged in machine learning, particularly in the study of Hopfield networks, Boltzmann machines, and ensemble learning, where learning processes have been analysed using tools from the statistical mechanics of disordered systems, stochastic stability, and replica theory [ 10 , 36 ] . In particular, the training dynamics and hyper-parameter optimisation of models such as Restricted Boltzmann Machines can be interpreted within frameworks that resemble partially annealed systems. In such frameworks, parameters and dynamical variables evolve at different effective temperatures or timescales [ 6 , 23 ] . This viewpoint establishes a connection between classical disordered systems and contemporary learning theory, providing new perspectives on glassy behaviour, optimisation landscapes, and the interpretability of trained models. The objective of this study is to provide a theoretical description of this intermediate regime by means of Guerra’s interpolation scheme, within both the RS frameworks. This study builds on the foundational research of Dotsenko et al [ 22 ] , investigating the role of the number of replicas in this context, which is naturally extended to take real values. The manuscript is organized as follows. After a general introduction of the Hopfield model in Sec. 2 , we