A Three-Threshold Learning Rule Approaches the Maximal Capacity of Recurrent Neural Networks.

Understanding the theoretical foundations of how memories are encoded and retrieved in neural populations is a central challenge in neuroscience. A popular theoretical scenario for modeling memory function is the attractor neural network scenario, whose prototype is the Hopfield model. The model simplicity and the locality of the synaptic update rules come at the cost of a poor storage capacity, compared with the capacity achieved with perceptron learning algorithms. Here, by transforming the perceptron learning rule, we present an online learning rule for a recurrent neural network that achieves near-maximal storage capacity without an explicit supervisory error signal, relying only upon locally accessible information. The fully-connected network consists of excitatory binary neurons with plastic recurrent connections and non-plastic inhibitory feedback stabilizing the network dynamics; the memory patterns to be memorized are presented online as strong afferent currents, producing a bimodal distribution for the neuron synaptic inputs. Synapses corresponding to active inputs are modified as a function of the value of the local fields with respect to three thresholds. Above the highest threshold, and below the lowest threshold, no plasticity occurs. In between these two thresholds, potentiation/depression occurs when the local field is above/below an intermediate threshold. We simulated and analyzed a network of binary neurons implementing this rule and measured its storage capacity for different sizes of the basins of attraction. The storage capacity obtained through numerical simulations is shown to be close to the value predicted by analytical calculations. We also measured the dependence of capacity on the strength of external inputs. Finally, we quantified the statistics of the resulting synaptic connectivity matrix, and found that both the fraction of zero weight synapses and the degree of symmetry of the weight matrix increase with the number of stored patterns.

A Lyapunov theory demonstrating a fundamental limit on the speed of systems consolidation.

The nervous system reorganizes memories from an early site to a late site, a commonly observed feature of learning and memory systems known as systems consolidation. Previous work has suggested learning rules by which consolidation may occur. Here, we provide conditions under which such rules are guaranteed to lead to stable convergence of learning and consolidation. We use the theory of Lyapunov functions, which enforces stability by requiring learning rules to decrease an energy-like (Lyapunov) function. We present the theory in the context of a simple circuit architecture motivated by classic models of learning in systems consolidation mediated by the cerebellum. Stability is only guaranteed if the learning rate in the late stage is not faster than the learning rate in the early stage. Further, the slower the learning rate at the late stage, the larger the perturbation the system can tolerate with a guarantee of stability. We provide intuition for this result by mapping the consolidation model to a damped driven oscillator system, and showing that the ratio of early-to late-stage learning rates in the consolidation model can be directly identified with the (square of the) oscillator's damping ratio. This work suggests the power of the Lyapunov approach to provide constraints on nervous system function.

Seminal plasma proteomics as putative biomarkers for male infertility diagnosis.

Male infertility represents a significant global public health issue that is currently emerging as a prominent research focus. Presently, laboratories adhere to the guidelines outlined by the World Health Organization (WHO) manuals for conducting routine semen analysis to diagnose male infertility. However, the accuracy of results in predicting sperm quality and fertility is limited because some individuals with a normal semen analysis report, an unremarkable medical history, and a physical examination may still experience infertility. As a result, the importance of employing more advanced techniques to investigate sperm function and male fertility in the treatment of male infertility and/or subfertility becomes apparent. The standard test for evaluating human semen has been improved by more complex tests that look at things like reactive oxygen species (ROS) levels, total antioxidant capacity (TAC), sperm DNA fragmentation levels, DNA compaction, apoptosis, genetic testing, and the presence and location of anti-sperm antibodies. Recent discoveries of novel biomarkers have significantly enriched our understanding of male fertility. Moreover, the notable biological diversity among samples obtained from the same individual complicates the efficacy of routine semen analysis. Therefore, unraveling the molecular mechanisms involved in fertilization is pivotal in expanding our understanding of factors contributing to male infertility. By understanding how these proteins work and what role they play in sperm activity, we can look at the expression profile in men who can't have children to find diagnostic biomarkers. This review examines the various sperm and seminal plasma proteins associated with infertility, as well as proteins that are either deficient or exhibit aberrant expression, potentially contributing to male infertility causes.

Reconstruction of an informative railway wheel defect signal from wheel-rail contact signals measured by multiple wayside sensors.

Wheel impact load detectors are widespread railway systems used for measuring the wheel-rail contact force. They usually measure the rail strain and convert it to force in order to detect high impact forces and corresponding detrimental wheels. The measured strain signal can also be used to identify the defect type and its severity. The strain sensors have a limited effective zone that leads to partial observation from the wheels. Therefore, wheel impact load detectors exploit multiple sensors to collect samples from different portions of the wheels. The discrete measurement by multiple sensors provides the magnitude of the force; however, it does not provide the much richer variation pattern of the contact force signal. Therefore, this paper proposes a fusion method to associate the collected samples to their positions over the wheel circumferential coordinate. This process reconstructs an informative signal from the discrete samples collected by multiple sensors. To validate the proposed method, the multiple sensors have been simulated by an ad hoc multibody dynamic software (VI-Rail), and the outputs have been fed to the fusion model. The reconstructed signal represents the contact force and consequently the wheel defect. The obtained results demonstrate considerable similarity between the contact force and the reconstructed defect signal that can be used for further defect identification.

Micromechanical analysis of brain's diffuse axonal injury.

Computational models are important tools which help researchers understand traumatic brain injury (TBI). A mechanistic multi-scale numerical approach is introduced to quantify diffuse axonal injury (DAI), the most important mechanism of TBI, induced by a mechanical insult at micro-scale regions of the white matter or voxels where fiber orientations are the same. Using the mechanical properties of a single axon with a viscoelastic constitutive relation and its functional failure in terms of electrophysiological impairment, a numerical 2D micro-level lattice method is implemented to directly analyze the percentage of injured axons in a voxel containing a bundle of axons all with the same orientation under biaxial stretches. Reference micro-injury maps are then developed with the input parameters based on the principal strain or stretch values and their direction with respect to axons, which provide the percentage of injured axons in the voxel of interest as the output. The methodology is independent of any statistical analyses of the accident data and medical reports to derive probabilistic injury risk curves for DAI. Avoiding a structurally detailed full finite element head model, this study proposes a micro-mechanical approach which considers the anatomical structure of neural axons in the white matter together with their mechanical properties using a numerical lattice method to analyze the brain's diffuse axonal injury. This work has the potential to help develop safer prevention tools and more effective diagnosis methods for DAI.

The Brain as an Efficient and Robust Adaptive Learner.

Understanding how the brain learns to compute functions reliably, efficiently, and robustly with noisy spiking activity is a fundamental challenge in neuroscience. Most sensory and motor tasks can be described as dynamical systems and could presumably be learned by adjusting connection weights in a recurrent biological neural network. However, this is greatly complicated by the credit assignment problem for learning in recurrent networks, e.g., the contribution of each connection to the global output error cannot be determined based only on locally accessible quantities to the synapse. Combining tools from adaptive control theory and efficient coding theories, we propose that neural circuits can indeed learn complex dynamic tasks with local synaptic plasticity rules as long as they associate two experimentally established neural mechanisms. First, they should receive top-down feedbacks driving both their activity and their synaptic plasticity. Second, inhibitory interneurons should maintain a tight balance between excitation and inhibition in the circuit. The resulting networks could learn arbitrary dynamical systems and produce irregular spike trains as variable as those observed experimentally. Yet, this variability in single neurons may hide an extremely efficient and robust computation at the population level.

Object similarity affects the perceptual strategy underlying invariant visual object recognition in rats.

In recent years, a number of studies have explored the possible use of rats as models of high-level visual functions. One central question at the root of such an investigation is to understand whether rat object vision relies on the processing of visual shape features or, rather, on lower-order image properties (e.g., overall brightness). In a recent study, we have shown that rats are capable of extracting multiple features of an object that are diagnostic of its identity, at least when those features are, structure-wise, distinct enough to be parsed by the rat visual system. In the present study, we have assessed the impact of object structure on rat perceptual strategy. We trained rats to discriminate between two structurally similar objects, and compared their recognition strategies with those reported in our previous study. We found that, under conditions of lower stimulus discriminability, rat visual discrimination strategy becomes more view-dependent and subject-dependent. Rats were still able to recognize the target objects, in a way that was largely tolerant (i.e., invariant) to object transformation; however, the larger structural and pixel-wise similarity affected the way objects were processed. Compared to the findings of our previous study, the patterns of diagnostic features were: (i) smaller and more scattered; (ii) only partially preserved across object views; and (iii) only partially reproducible across rats. On the other hand, rats were still found to adopt a multi-featural processing strategy and to make use of part of the optimal discriminatory information afforded by the two objects. Our findings suggest that, as in humans, rat invariant recognition can flexibly rely on either view-invariant representations of distinctive object features or view-specific object representations, acquired through learning.