External Stimuli on Neural Networks: Analytical and Numerical Approaches

Based on the behavior of living beings, which react mostly to external stimuli, we introduce a neural-network model that uses external patterns as a fundamental tool for the process of recognition. In this proposal, external stimuli appear as an additional field, and basins of attraction, representing memories, arise in accordance with this new field. This is in contrast to the more-common attractor neural networks, where memories are attractors inside well-defined basins of attraction. We show that this procedure considerably increases the storage capabilities of the neural network; this property is illustrated by the standard Hopfield model, which reveals that the recognition capacity of our model may be enlarged, typically, by a factor 102. The primary challenge here consists in calibrating the influence of the external stimulus, in order to attenuate the noise generated by memories that are not correlated with the external pattern. The system is analyzed primarily through numerical simulations. However, since there is the possibility of performing analytical calculations for the Hopfield model, the agreement between these two approaches can be tested—matching results are indicated in some cases. We also show that the present proposal exhibits a crucial attribute of living beings, which concerns their ability to react promptly to changes in the external environment. Additionally, we illustrate that this new approach may significantly enlarge the recognition capacity of neural networks in various situations; with correlated and non-correlated memories, as well as diluted, symmetric, or asymmetric interactions (synapses). This demonstrates that it can be implemented easily on a wide diversity of models.

The Sakaguchi-Kuramoto Model in Presence of Asymmetric Interactions That Break Rotational Symmetry.

The celebrated Kuramoto model provides an analytically tractable framework to study spontaneous collective synchronization, and comprises globally coupled limit-cycle oscillators interacting symmetrically with one another. The Sakaguchi-Kuramoto model is a generalization of the basic model that considers the presence of a phase-lag parameter in the interaction, thereby making it asymmetric between oscillator pairs. Here, we consider a further generalization, by adding an interaction that breaks the rotational symmetry of the model. The highlight of our study is the unveiling of a very rich phase diagram comprising both oscillatory and non-oscillatory synchronized phases as well as an incoherent phase: There are regions of two-phase as well as an interesting and hitherto unexplored three-phase coexistence arising from asymmetric interactions in our model.

Survival of the weakest in non-transitive asymmetric interactions among strains of E. coli

Hierarchical organization in ecology, whereby interactions are nested in a manner that leads to a dominant species, naturally result in the exclusion of all but the dominant competitor. Alternatively, non-hierarchical competitive dynamics, such as cyclical interactions, can sustain biodiversity. Here, we designed a simple microbial community with three strains of E. coli that cyclically interact through (i) the inhibition of protein production, (ii) the digestion of genomic DNA, and (iii) the disruption of the cell membrane. We find that intrinsic differences in these three major mechanisms of bacterial warfare lead to an unbalanced community that is dominated by the weakest strain. We also use a computational model to describe how the relative toxin strengths, initial fractional occupancies, and spatial patterns affect the maintenance of biodiversity. The engineering of active warfare between microbial species establishes a framework for exploration of the underlying principles that drive complex ecological interactions.

Cherry Picking: Consumer Choices in Swarm Dynamics, Considering Price and Quality of Goods

This paper proposes a further development of the mathematical theory of swarms to behavioral dynamics of social and economic systems, with an application to the modeling of price series in a market. The complexity features of the system are properly described by modeling the asymmetric interactions between buyers and sellers, specifically considering the so-called cherry picking phenomenon, by which not only prices but also qualities are considered when buying a good. Finally, numerical simulations are performed to depict the predictive ability of the model and to show interesting emerging behaviors, as the coordination of buyers and their division in endogenous clusters.

Programming van der Waals interactions with complex symmetries into microparticles using liquid crystallinity

Asymmetric interactions such as entropic (e.g., encoded by nonspherical shapes) or surface forces (e.g., encoded by patterned surface chemistry or DNA hybridization) provide access to functional states of colloidal matter, but versatile approaches for engineering asymmetric van der Waals interactions have the potential to expand further the palette of materials that can be assembled through such bottom-up processes. We show that polymerization of liquid crystal (LC) emulsions leads to compositionally homogeneous and spherical microparticles that encode van der Waals interactions with complex symmetries (e.g., quadrupolar and dipolar) that reflect the internal organization of the LC. Experiments performed using kinetically controlled probe colloid adsorption and complementary calculations support our conclusion that LC ordering can program van der Waals interactions by ~20 kBT across the surfaces of microparticles. Because diverse LC configurations can be engineered by confinement, these results provide fresh ideas for programming van der Waals interactions for assembly of soft matter.