Computational Core Research Articles

On reduction and normalization in the computational core

AbstractWe study the reduction in a $\lambda$ -calculus derived from Moggi’s computational one, which we call the computational core. The reduction relation consists of rules obtained by orienting three monadic laws. Such laws, in particular associativity and identity, introduce intricacies in the operational analysis. We investigate the central notions of returning a value versus having a normal form and address the question of normalizing strategies. Our analysis relies on factorization results.

Dynamics and control simulation of railway virtual coupling

ABSTRACT This paper developed a parallel computing architecture for high-fidelity virtual coupling simulations. Multi-body train dynamics models considered various nonlinear components including wheel-rail contact, suspensions, and inter-vehicle connections. A virtual coupling controller was developed which can be implemented under various train-to-train communication topologies. The controller also allows existing trains to leave the platoon and new trains to merge into the platoon without re-designing the controller. The parallel computing architecture is also scalable and not limited by: the number of vehicles in each train; the number of trains in each train platoon and the topology of train-to-train communications. A case study by simulating a three-train (18 vehicles in total) platoon on a real-world track section was conducted. The results show that, by using 19 computer cores, parallel computing speed is nearly twice as fast as real-time. Parallel computing is about 17 times faster than serial computing. The results also show that the maximum spacing errors of the follower trains were about 0.22 m. Dynamics results such as wheel-rail contact forces, suspension forces, carbody vibrations and inter-vehicle forces were obtained; these results can be used to conduct system assessments in terms of passenger ride comfort, mechanical wear, etc.

Reinforcement learning with associative or discriminative generalization across states and actions: fMRI at 3T and 7T.

The model‐free algorithms of “reinforcement learning” (RL) have gained clout across disciplines, but so too have model‐based alternatives. The present study emphasizes other dimensions of this model space in consideration of associative or discriminative generalization across states and actions. This “generalized reinforcement learning” (GRL) model, a frugal extension of RL, parsimoniously retains the single reward‐prediction error (RPE), but the scope of learning goes beyond the experienced state and action. Instead, the generalized RPE is efficiently relayed for bidirectional counterfactual updating of value estimates for other representations. Aided by structural information but as an implicit rather than explicit cognitive map, GRL provided the most precise account of human behavior and individual differences in a reversal‐learning task with hierarchical structure that encouraged inverse generalization across both states and actions. Reflecting inference that could be true, false (i.e., overgeneralization), or absent (i.e., undergeneralization), state generalization distinguished those who learned well more so than action generalization. With high‐resolution high‐field fMRI targeting the dopaminergic midbrain, the GRL model's RPE signals (alongside value and decision signals) were localized within not only the striatum but also the substantia nigra and the ventral tegmental area, including specific effects of generalization that also extend to the hippocampus. Factoring in generalization as a multidimensional process in value‐based learning, these findings shed light on complexities that, while challenging classic RL, can still be resolved within the bounds of its core computations.

6-qubit optimal Clifford circuits

Clifford group lies at the core of quantum computation—it underlies quantum error correction, its elements can be used to perform magic state distillation and they form randomized benchmarking protocols, Clifford group is used to study quantum entanglement, and more. The ability to utilize Clifford group elements in practice relies heavily on the efficiency of their circuit-level implementation. Finding short circuits is a hard problem; despite Clifford group being finite, its size grows quickly with the number of qubits n, limiting known optimal implementations to n = 4 qubits. For n = 6, the number of Clifford group elements is about 2.1 × 1023. In this paper, we report a set of algorithms, along with their C++ implementation, that implicitly synthesize optimal circuits for all 6-qubit Clifford group elements by storing a subset of the latter in a database of size 2.1TB (1kB = 1024B). We demonstrate how to extract arbitrary optimal 6-qubit Clifford circuit in 0.0009358 and 0.0006274 s using consumer- and enterprise-grade computers (hardware) respectively, while relying on this database. We use this implementation to establish a new example of quantum advantage by Clifford circuits over CNOT gate circuits and find optimal Clifford 2-designs for up to 4 qubits.

Phase Change Memories in Smart Sensing Solutions for Structural Health Monitoring

Smart devices for structural health monitoring provide edge computing capabilities to reduce wireless transmission and, thus, power consumption. Although effective algorithms have been proposed in the last few decades, traditional microcontrollers require heavy data flow between the memory and the central processing unit that involves a considerable fraction of the total energy consumption. Phase change memory has recently emerged as an attractive solution in the field of resistive nonvolatile memory for analog in-memory computing, which is a valid approach to avoid data being conveyed among distinct elaboration units. However, it has never been envisaged in structural health monitoring applications. As this technology is still in an embryonic state, several challenges related to nonlinearities and nonidealities of the memory elements and the energy expenditure related to the memory reprogramming process may undermine its usage. In this paper, the application of a novel identification approach for civil infrastructures is investigated using phase change memories. The main computational core of the presented algorithm, consisting of one-dimensional convolutions, is particularly suitable for implementations involving analog in-memory computing, thus showing the great potential of this technology for structural health monitoring applications. The test unit is an embedded phase change memory provided by STMicroelectronics and designed in 90-nm smart power bipolar complementary metal-oxide-semiconductor (CMOS)-double-diffused metal-oxide-semiconductor (DMOS) technology with a Ge-rich Ge-Sb-Te alloy for automotive applications. Experimental results obtained for a viaduct of an Italian motorway support the efficacy of the method. Moreover, the influence of nonidealities on the outcomes of damage identification based on both dynamic and quasi-static structural parameters is examined.

Cognitive and Epistemic Spaces in the CSTA Computer Science Learning Standards

This paper presents the findings of close reading of the Computer Science Teachers Association (CSTA) Computer Science Learning Standards for grades K-12. This document is influential because it has been adopted by at least five of the United States and has served as an inspiration for several more. Analyses of the document structure as well as of verb usage, word frequencies, and word pair networks for the language of the standards themselves as well as the clarification text included in the document for each standard were conducted using tidytext tools built for the R programming language. These analyses revealed a high level of internal coherence in this learning standards document, as well as a wide epistemic space for the teaching and learning of core computer science concepts and practices by K-12 students and their teachers. Implications for student learning, curriculum design, and computer science teacher preparation were discussed.

Fast Estimation of Outcome Probabilities for Quantum Circuits

We present two classical algorithms for the simulation of universal quantum circuits on $n$ qubits constructed from $c$ instances of Clifford gates and $t$ arbitrary-angle $Z$-rotation gates such as $T$ gates. Our algorithms complement each other by performing best in different parameter regimes. The $\tt{Estimate}$ algorithm produces an additive precision estimate of the Born rule probability of a chosen measurement outcome with the only source of run-time inefficiency being a linear dependence on the stabilizer extent (which scales like $\approx 1.17^t$ for $T$ gates). Our algorithm is state-of-the-art for this task: as an example, in approximately $13$ hours (on a standard desktop computer), we estimated the Born rule probability to within an additive error of $0.03$, for a $50$-qubit, $60$ non-Clifford gate quantum circuit with more than $2000$ Clifford gates. Our second algorithm, $\tt{Compute}$, calculates the probability of a chosen measurement outcome to machine precision with run-time $O(2^{t-r} t)$ where $r$ is an efficiently computable, circuit-specific quantity. With high probability, $r$ is very close to $\min \{t, n-w\}$ for random circuits with many Clifford gates, where $w$ is the number of measured qubits. $\tt{Compute}$ can be effective in surprisingly challenging parameter regimes, e.g., we can randomly sample Clifford+$T$ circuits with $n=55$, $w=5$, $c=10^5$ and $t=80$ $T$ gates, and then compute the Born rule probability with a run-time consistently less than $10$ minutes using a single core of a standard desktop computer. We provide a C+Python implementation of our algorithms and benchmark them using random circuits, the hidden shift algorithm and the quantum approximate optimization algorithm (QAOA).

Differential Dopamine Receptor-Dependent Sensitivity Improves the Switch Between Hard and Soft Selection in a Model of the Basal Ganglia.

The problem of selecting one action from a set of different possible actions, simply referred to as the problem of action selection, is a ubiquitous challenge in the animal world. For vertebrates, the basal ganglia (BG) are widely thought to implement the core computation to solve this problem, as its anatomy and physiology are well suited to this end. However, the BG still display physiological features whose role in achieving efficient action selection remains unclear. In particular, it is known that the two types of dopaminergic receptors (D1 and D2) present in the BG give rise to mechanistically different responses. The overall effect will be a difference in sensitivity to dopamine, which may have ramifications for action selection. However, which receptor type leads to a stronger response is unclear due to the complexity of the intracellular mechanisms involved. In this study, we use an existing, high-level computational model of the BG, which assumes that dopamine contributes to action selection by enabling a switch between different selection regimes, to predict which of D1 or D2 has the greater sensitivity. Thus, we ask, Assuming dopamine enables a switch between action selection regimes in the BG, what functional sensitivity values would result in improved action selection computation? To do this, we quantitatively assessed the model's capacity to perform action selection as we parametrically manipulated the sensitivity weights of D1 and D2. We show that differential (rather than equal) D1 and D2 sensitivity to dopaminergic input improves the switch between selection regimes during the action selection computation in our model. Specifically, greater D2 sensitivity compared to D1 led to these improvements.

Multiple same-level and telescoping nesting in GFDL's dynamical core

Abstract. Two-way multiple same-level and telescoping grid nesting capabilities are implemented in the Geophysical Fluid Dynamics Laboratory (GFDL)'s Finite-Volume Cubed-Sphere Dynamical Core (FV3). Simulations are performed within GFDL's System for High-resolution modeling for Earth-to-Local Domains (SHiELD) using global and regional multiple nest configurations. Results show that multiple same-level and multi-level telescoping nests were able to capture various weather events in greater details by resolving smaller-scale flow structures. Two-way updates do not introduce numerical errors in their corresponding parent grids where the nests are located. The cases of Hurricane Laura's landfall and an atmospheric river in California were found to be more intense with increased levels of telescoping nesting. All nested grids run concurrently, and adding additional nests with computer cores to a setup does not degrade the computational performance nor increase the simulation run time if the cores are optimally distributed among the grids.

Parallel-in-time simulation of biofluids

We extend Parareal, a parallel-in-time method that alternates between a serial sweep and a parallel sweep, to simulate the fluid flow around bio-inspired, dynamic structures over a period of time. Our main contributions include demonstrating the applicability of Parareal to the simulation of biofluids and developing novel solvers for the serial sweeps of Parareal. We propose to construct non-intrusive solvers by extrapolating a parametrized family of existing solvers. Compared to the existing solvers, they either allow the use of larger time steps, have a higher order of accuracy in time, or both. They are also straightforward to implement and parallelize. Numerical results show that when the number of biological structures is small or the number of computer cores employed is sufficiently large, the proposed variant of Parareal can achieve a significantly higher parallel speedup than the more commonly used spatial parallelization.

ViP: A Hierarchical Parallel Vision Processor for Hybrid Vision Chip

Nowadays, the vision chip bridging sensing and processing has been extensively employed in high-speed image processing, owing to its excellent performance, low power consumption, and economical cost. However, there is a dilemma in designing processors to support conventional computer vision algorithms and neural networks since the two algorithms have a non-trivial trade-off in proposing a unified architecture. By analyzing computation properties, we propose a novel hierarchical parallel vision processor (ViP) for hybrid vision chips to accelerate both traditional computer vision (CV) and neural network (NN). The ViP architecture includes three parallelism levels: PE for pixel-centric, computing core (CC) for block, and vision core (VC) for global. PEs contain dedicated computing units and data paths for convolution operations without degrading its flexibility. Each CC is driven by customized SIMD instructions and can be dynamically connected for meeting block parallelism requirements. ViP is fabricated in 65nm CMOS technology and achieves a peak performance of 614.4 GOPS and an energy efficiency of 640 GOPS/W at 200 MHz clock frequency. Notably, several experiments on CV and NN are performed, illustrating an ultra-low latency in executing hybrid algorithms.

The Jacobson radical for an inconsistency predicate

As a form of the Axiom of Choice about relatively simple structures (posets), Hausdorff’s Maximal Chain Principle appears to be little amenable to computational interpretation. This received view, however, requires revision: maximal chains are more reminiscent of maximal ideals than it seems at first glance. The latter live in richer algebraic structures (rings), and thus are readier to be put under computational scrutiny. Exploiting this, and of course the analogy between maximal chains and maximal ideals, the concept of Jacobson radical carries over from a ring to an arbitrary set with an abstract inconsistency predicate: that is, a distinguished monotone family of finite subsets. All this makes possible not only to generalise Hausdorff’s principle, but also to express it as a syntactical conservation theorem. The latter, which encompasses the desired computational core of Hausdorff’s principle, is obtained by a generalised inductive definition. The over-all setting is constructive set theory.

An Artificial Visual System for Motion Direction Detection Based on the Hassenstein–Reichardt Correlator Model

The perception of motion direction is essential for the survival of visual animals. Despite various theoretical and biophysical investigations that have been conducted to elucidate directional selectivity at the neural level, the systemic mechanism of motion direction detection remains elusive. Here, we develop an artificial visual system (AVS) based on the core computation of the Hassenstein–Reichardt correlator (HRC) model for global motion direction detection. With reference to the biological investigations of Drosophila, we first describe a local motion-sensitive, directionally detective neuron that only responds to ON motion signals with high pattern contrast in a particular direction. Then, we use the full-neurons scheme motion direction detection mechanism to detect the global motion direction based on our previous research. The mechanism enables our AVS to detect multiple directions in a two-dimensional view, and the global motion direction is inferred from the outputs of all local motion-sensitive directionally detective neurons. To verify the reliability of our AVS, we conduct a series of experiments and compare its performance with the time-considered convolution neural network (CNN) and the EfficientNetB0 under the same conditions. The experimental results demonstrated that our system is reliable in detecting the direction of motion, and among the three models, our AVS has better motion direction detection capabilities.

Longitudinal form Factors of 25Mg with Bohr-Mottelson Model Using Large-scale Shell Model

The longitudinal electron scattering form factors in large scale SDPF-shell for 25Mg with SDPF now as effective interaction have been discussed using Bohr-Mottelson model to computation core polarization (CP). The calculation are done with the help of NuShell code. To determine the single-particle wave function, the SKyrme-Hartree Fock potential was used to compute the matrix elements' radial wave function containing a single particle. Bohr-Mottelson model is used in these calculation. The result show good agreement with experimental data for 25Mg nucleus.

BIODICA: a computational environment for Independent Component Analysis.

We developed BIODICA, an integrated computational environment for application of Independent Component Analysis (ICA) to bulk and single-cell molecular profiles, interpretation of the results in terms of biological functions and correlation with metadata. The computational core is the novel Python package stabilized-ica which provides interface to several ICA algorithms, a stabilization procedure, meta-analysis and component interpretation tools. BIODICA is equipped with a user-friendly graphical user interface, allowing non-experienced users to perform the ICA-based omics data analysis. The results are provided in interactive ways, thus facilitating communication with biology experts. BIODICA is implemented in Java, Python and JavaScript. The source code is freely available on GitHub under the MIT and the GNU LGPL licenses. BIODICA is supported on all major operating systems. https://sysbio-curie.github.io/biodica-environment/.

An 8-Bit in Resistive Memory Computing Core With Regulated Passive Neuron and Bitline Weight Mapping

The rapid development of artificial intelligence (AI) and Internet of Things (IoT) increase the requirement for edge computing with low power and relatively high processing speed devices. The computing-in-memory (CIM) schemes based on emerging resistive nonvolatile memory (NVM) show great potential in reducing the power consumption for AI computing. However, the inconsistency of the NVM may significantly degenerate the performance of the neural network. In this article, we propose a low power resistive RAM (RRAM)-based CIM core to not only achieve high computing efficiency but also greatly enhance the robustness by bit line (BL) regulator and BL weight mapping algorithm. The simulation results show that the power consumption of our proposed 8-bit CIM core is only 12.6 mW ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$256\times 256$ </tex-math></inline-formula> at 8b). The spurious-free dynamic range (SFDR) and signal to noise and distortion ratio (SNDR) of the CIM core achieve 62.64 and 45.92 dB, respectively. The proposed BL weight mapping scheme improves the top-1 accuracy by 2.46% and 3.47% for AlexNet and VGG16 on ImageNet Large Scale Visual Recognition Competition 2012 (ILSVRC 2012) in 8-bit mode, respectively.

Humans can navigate complex graph structures acquired during latent learning

Humans appear to represent many forms of knowledge in associative networks whose nodes are multiply connected, including sensory, spatial, and semantic. Recent work has shown that explicitly augmenting artificial agents with such graph-structured representations endows them with more human-like capabilities of compositionality and transfer learning. An open question is how humans acquire these representations. Previously, it has been shown that humans can learn to navigate graph-structured conceptual spaces on the basis of direct experience with trajectories that intentionally draw the network contours (Schapiro, Kustner, & Turk-Browne, 2012; Schapiro, Turk-Browne, Botvinick, & Norman, 2016), or through direct experience with rewards that covary with the underlying associative distance (Wu, Schulz, Speekenbrink, Nelson, & Meder, 2018). Here, we provide initial evidence that this capability is more general, extending to learning to reason about shortest-path distances across a graph structure acquired across disjoint experiences with randomized edges of the graph - a form of latent learning. In other words, we show that humans can infer graph structures, assembling them from disordered experiences. We further show that the degree to which individuals learn to reason correctly and with reference to the structure of the graph corresponds to their propensity, in a separate task, to use model-based reinforcement learning to achieve rewards. This connection suggests that the correct acquisition of graph-structured relationships is a central ability underlying forward planning and reasoning, and may be a core computation across the many domains in which graph-based reasoning is advantageous.

Внедрение методов суперкомпьютерной оптимизации в процесс проектирования водометных движителей

Object and purpose of research. This study discusses fast boats and waterjet propulsors. The purpose is to design a waterjet impeller shape to required conditions by means of supercomputer-based optimization with further validation of results at sea trials. Materials and methods. The optimization process is governed by a Russian software package pSeven. The computation core is Star CCM+ (Siemens). Parametric 3D model of the impeller is generated in BladePlus software (in-house KSRC development). Hydrodynamic parameters of waterjet are calculated as per CFD methods. Viscous flow parameters are found through the control-volume solution of unsteady Reynolds equations (URANS) closed by the bi-parametric semi-empirical turbulence model. Main results. The optimization studies yielded new shapes for impellers, stators and flow parts of waterjets. These solu-tions are more efficient than the prototype thanks to greater thrust of the propulsion system and lower resistance of the flow part without prejudice to good cavitation performance. Calculated hydrodynamic parameters of waterjets and the estimates for achievable speed of fast boats have been confirmed by the sea trials. Conclusion. Supercomputer optimization ensured successful beginning of the development of Russian waterjet family for fast boats not inferior to universally recognized market leaders in terms of performance parameters.

Can the jigsaw puzzle model of protein folding re-assemble a hydrophobic core?

According to the "jigsaw puzzle" model of protein folding, the isomorphism between sequence and structure is substantially determined by the specific geometry of side-chain interactions, within the protein interior. In this work, we have attempted to predict the hydrophobic core of cyclophilin (LdCyp) from Leishmania donovani, utilizing a surface complementarity function, which selects for high goodness of fit between hydrophobic side-chain surfaces, rather in the manner of assembling a three-dimensional jigsaw puzzle. The computational core prediction method implemented here has been tried on two distinct scenarios, on the LdCyp polypeptide chain with native non-core residues and all core residues initially set to alanine, on a poly-glycine polypeptide chain. Molecular dynamics simulations appeared to indicate partial destabilization of the two designed sequences. However, experimental characterization of the designed sequences by circular dichroism (CD) spectroscopy and denaturant (GdmCl) induced unfolding, demonstrated disordered proteins. Stepwise reconstruction of the designed cores by cumulative sequential mutations identified the specific mutation (M122L) as primarily responsible for fold collapse and all design objectives were achieved upon rectifying this mutation. In summary, the study demonstrates regions of the core to contain highly specific (jigsaw puzzle-like) interactions sensitive to any perturbations and a predictive algorithm to identify such regions. A mutation within the core has been identified which exercises an inordinate influence on the global fold, reminiscent of metamorphic proteins. In addition, the computational procedure could predict substantial regions of the core (given main-chain coordinates) without any reference to non-core residues.

An intelligent optimization method of reload core loading pattern and its application

Reload core loading pattern design is the key content related to safety, economy and flexibility of the operation after refueling in nuclear power plant. The optimization design of the reload loading pattern is a very short time requirements, great difficulty, high experience requirements and repeatable work. According to these above rigorous requirements, based on statistical learning and random sampling, this study proposes an intelligent method of reload core loading pattern optimization——Probability Sampling Loading Pattern Optimization (PSLPO). This method intelligently extracts the characteristic law in the optimization space of the reload core loading pattern by statistical learning method to establish the probability model. And the probability model is embedded into PSLPO by the random sampling method. Finally, PSLPO method is constantly updated and evolved in the optimization process by intelligent means such as dynamic adjustment weight factor, probability model adaptive adjustment mechanism and combined simulated annealing design, which realizes the improvement of the optimization ability of the algorithm continuously. According to the actual reload design verification for cycle 5 of Hainan nuclear power plant unit 2, it is shown that the recommended pattern of PSLPO is better than manual search pattern in the aspect of cycle length and power distribution, and the method can intelligently give multiple design patterns, which users can have many choices. Through the optimization evaluation with a single computer core, the optimization efficiency of the method can meet the needs of core reload design. The global optimization ability can be further improved through parallelization and algorithm parameter optimization.