How does carbon emissions trading scheme affect the investment efficiency of companies? Evidence from China.

Carbon emissions trading scheme (ETS) will change the internal and external operating environment of related companies, increase new carbon reduction costs, and may affect their investment behavior and decision-making. This study constructs the policy effect evaluation model to evaluate the net effect of China's ETS pilot policy on the investment efficiency of related companies in high-carbon industries. This study indicates that: first, China's ETS pilot policy has reduced the inefficient investment of related companies by 11.40% during the sample period. Second, innovation ability plays an enhanced moderating effect in the inhibitory effect of China's ETS pilot policy on the inefficient investment of related companies in the short term, but the effect is weakened in the long-term. Third, the policy effect is relatively obvious for non-state-owned companies and companies with more institutional shareholdings. This research has important reference value for the continuous improvement of China's national carbon market, and the steady transformation from simple investment-driven to investment-efficiency-driven economic development in China.

Evaluating the efficiency of online course resource allocation in universities of China.

Curriculum is the core element of talent cultivation in universities, and its quality directly determines that of the talent cultivation. However, the efficiency of online course resource allocation in universities of China is unacceptable, and the corresponding driving mechanisms are unclear. Based on the data of iCourses platform for a top university (H University) of China, this paper develops the DEA-Malmquist index and Tobit regression model to evaluate the efficiency of university online course resource allocation and explore its main influencing factors. The empirical results indicate that, first of all, humanities and social sciences outperform natural sciences in the resource allocation efficiency of online courses. Compared with non-national premium online courses, the resource investment redundancy of national premium online courses appears relatively lower. Secondly, from the perspective of temporal changes, the total factor productivity of online course resource allocation generally has a downward trend. The innovation effect of technological progress is more significant, while the catch-up effect of technical efficiency is clearly insufficient. Scale efficiency is the main factor hindering its improvement. Finally, in terms of the factors affecting the efficiency of online course resource allocation, teachers' research guidance capability, teaching research skills, and the number of courses offered have significantly positive impact.

An efficient ANN SoC for detecting Alzheimer's disease based on recurrent computing.

Alzheimer's Disease (AD) is an irreversible, degenerative condition that, while incurable, can have its progression slowed or impeded. While there are numerous methods utilizing neural networks for AD detection, there is a scarcity of High-performance AD detection chips. Moreover, excessively complex neural networks are not conducive to on-chip implementation and clinical applications. This study addresses the challenges of high misdiagnosis rates and significant hardware costs inherent in traditional AD detection techniques. A novel and efficient AD detection framework based on a recurrent computational strategy is proposed. The framework harnesses an Artificial Neural Network (ANN) embedded within a System on Chip (SoC) to perform sophisticated Electroencephalogram (EEG) analysis. The approach began by employing a reduced IEEE754 single-precision encoding method to hardware-encode the preprocessed EEG data, thereby minimizing the memory storage area. Next, data remapping techniques were utilized to ensure the continuity of the input data read addresses and reduce the memory access pressure during neural network computations. Subsequently, hierarchical and Processing Element (PE) reuse technologies were leveraged to perform the multiply-accumulate operations of the ANN. Finally, a step function was chosen to establish binary classification circuits dedicated to AD detection. Experimental results indicate that the optimized SoC achieves a 70 % reduction in area and a 50 % reduction in power consumption compared to traditional designs. For various neural network models, the detection model proposed in this paper incurs less overhead, with a training speed 3 to 4 times faster than other traditional models, and a high accuracy rate of 98.53 %.

Cinnamaldehyde targets SarA to enhance ß-lactam antibiotic activity against methicillin-resistant Staphylococcus aureus.

Methicillin-resistant Staphylococcus aureus (MRSA) is a current global public health problem due to its increasing resistance to the most recent antibiotic therapies. One critical approach is to develop ways to revitalize existing antibiotics. Here, we show that the phytogenic compound cinnamaldehyde (CIN) and ß-lactam antibiotic combinations can functionally synergize and resensitize clinical MRSA isolates to ß-lactam therapy and inhibit MRSA biofilm formation. Mechanistic studies indicated that the CIN potentiation effect on ß-lactams was primarily the result of inhibition of the mecA expression by targeting the staphylococcal accessory regulator sarA. CIN alone or in combination with ß-lactams decreased sarA gene expression and increased SarA protein phosphorylation that impaired SarA binding to the mecA promoter element and downregulated virulence genes such as those encoding biofilm, α-hemolysin, and adhesin. Perturbation of SarA-mecA binding thus interfered with PBP2a biosynthesis and this decreased MRSA resistance to ß-lactams. Furthermore, CIN fully restored the anti-MRSA activities of ß-lactam antibiotics in vivo in murine models of bacteremia and biofilm infections. Together, our results indicated that CIN acts as a ß-lactam adjuvant and can be applied as an alternative therapy to combat multidrug-resistant MRSA infections.

Thermal Decomposition Mechanism of P(DAC-AM) with Serial Cationicity and Intrinsic Viscosity.

The thermal decomposition of the thermodynamic, kinetic and mechanisms of copolymer P(DAC-AM) samples with serial cationicity and intrinsic viscosity ([η]), and the control samples of homopolymer PAM and PDAC, were studied and analyzed using TG, DSC, FTIR. The results of the thermal decomposition thermodynamics confirmed that the thermal decomposition processes of the serial P(DAC-AM) samples and the two control samples could be divided into two stages. It was found that the processes of the copolymer P(DAC-AM) samples were not a simple superposition of those of homopolymers, whose monomers had composed the unit structures of the copolymer, but there were interactions between the two suspension groups. The results of thermal decomposition kinetics showed that the apparent activation energy (E) of the thermal decomposition process of all polymer samples had different varying trends in the terms of weight-loss rate (α). The reaction order (n) of the thermal decomposition of P(DAC-AM) in Stage I and II was close to 1, but in the former and the latter it tended to be 2 and 0.5, respectively. Finally, the thermal decomposition mechanism of copolymer P(DAC-AM) samples was discussed. The above research could not only fill in the knowledge vacancy of the thermal decomposition of the thermodynamic, kinetic and mechanisms of P(DAC-AM), but could also lay a foundation for the study of thermal decomposition mechanisms of the other types of polymers, including cationic polymers.

Tobramycin-resistant small colony variant mutant of Salmonella enterica serovar Typhimurium shows collateral sensitivity to nitrofurantoin.

The increasing antibiotic resistance poses a significant global health challenge, threatening our ability to combat infectious diseases. The phenomenon of collateral sensitivity, whereby resistance to one antibiotic is accompanied by increased sensitivity to another, offers potential avenues for novel therapeutic interventions against infections unresponsive to classical treatments. In this study, we elucidate the emergence of tobramycin (TOB)-resistant small colony variants (SCVs) due to mutations in the hemL gene, which render S. Typhimurium more susceptible to nitrofurantoin (NIT). Mechanistic studies demonstrate that the collateral sensitivity in TOB-resistant S. Typhimurium SCVs primarily stems from disruptions in haem biosynthesis. This leads to dysfunction in the electron transport chain (ETC) and redox imbalance, ultimately inducing lethal accumulation of reactive oxygen species (ROS). Additionally, the upregulation of nfsA/B expressions facilitates the conversion of NIT prodrug into its active form, promoting ROS-mediated bacterial killing and contributing to this collateral sensitivity pattern. Importantly, alternative NIT therapy demonstrates a significant reduction of bacterial load by more than 2.24-log10 cfu/g in the murine thigh infection and colitis models. Our findings corroborate the collateral sensitivity of S. Typhimurium to nitrofurans as a consequence of evolving resistance to aminoglycosides. This provides a promising approach for treating infections due to aminoglycoside-resistant strains.

Copolyester toughened poly(lactic acid) biodegradable material prepared by in situ formation of polyethylene glycol and citric acid.

Polylactic acid (PLA) is a high-modulus, high-strength bio-based thermoplastic polyester with good biodegradability, which is currently a promising environmentally friendly material. However, its inherent brittleness has hindered its widespread use. In this study, we reported a simple and non-toxic strategy for toughening PLA, using biodegradable materials such as polyethylene glycol (PEG) and citric acid (CA) as precursors. Through reactive melt blending with PLA, PEG and CA form PEGCA copolyesters in situ during blending. At the same time, CA can react with PLA and PEG, forming a copolyester structure at the interface of the two phases, improving the interfacial compatibility between PEG and PEGCA with PLA. Fourier transform infrared spectroscopy confirms this. Experimental results show that when the content of PEG/CA reaches 15% (PLA/PEG/CA-15%) in the blends, the impact strength of the blend was 4.47 kJ m-2, and the maximum elongation at break was as high as 360.1%, which were about 2 and 44 times higher than those of pure PLA, respectively. Moreover, the tensile strength was still maintained at the level of 70%. This work can expand the application of PLA in food packaging and medical supplies.

How to improve the energy-saving performance of China's transport sector? An input-output perspective.

The transport sector proves a major energy consumer in China, but improving energy-saving performance in China's provincial transport sector from the lifecycle perspective remains unresolved. Thus, this study employs the environmentally extended multi-region input-output (MRIO) method, structural path analysis, and the newest MRIO table of China from 2017, to investigate how to improve the energy-saving performance from final demand structure, supply chain, and pathway perspectives. The relevant results are threefold. (1) Regarding the final demand structure level, the embodied energy consumption of China's transport sector is predominantly driven by investment from the production side, while that of the consumption side is primarily caused by exports. (2) At the supply chain level, production-side embodied energy consumption primarily occurs along a three-echelon supply chain, while that from the consumption side mostly occurs via a two-echelon supply chain. (3) At the pathway level, the production-side energy-saving performance of China's provincial transport sector is dominated by two pathways along the construction sector, including transport sector → construction sector → final demands, and transport sector → intermediate inputs → construction sector → final demands, while that of the consumption side is chiefly determined by three pathways along internal transportation chains.

Molecular characteristic of mcr-1 gene in Escherichia coli from aquatic products in Guangdong, China.

OBJECTIVES: Aquatic ecosystems serve as a dissemination pathway and a reservoir of both antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs). This study aimed to determine the prevalence of colistin-resistant mcr-like genes in Enterobacteriales in aquatic products, which may be contribute to the transfer of ARGs in water environments. METHODS: The mcr-1-positive Escherichia coli were recovered from 123 freshwater fish and 34 cultured crocodile cecum samples from 10 farmers' markets in Guangdong, China. Minimum inhibitory concentration (MIC) was determined using the agar dilution method. Genotyping was performed using pulsed-field gel electrophoresis (PFGE) and multi-locus sequence typing (MLST). Conjugation assay was carried out to investigate the transferability of mcr-1. Genomic information was obtained by whole genome sequencing (WGS) and bioinformatic analysis. RESULTS: Forty-four mcr-1 positive isolates showed co-resistance to tetracycline, trimethoprim/sulfamethoxazole, and gentamicin, while they were all sensitive to tigecycline, meropenem, and amikacin. They were typed into sixteen PFGE clusters. ST10 and ST117 were the most popular sequence types, followed by ST1114. S1-PFGE verified the presence of the mcr-1 gene on plasmids in sizes of ∼60 kb (n = 1) and ∼240 kb (n = 3). Whole genome sequencing-based analysis identified mcr-1 integrated in IncHI2 plasmid (n = 3), IncI2 plasmid (n = 2), and bacterial chromosome in two copies (n = 1). In addition to mcr-1, they carried several other antibiotic resistance genes, such as blaCTX-M-14, fosA3, and aac(6')-Ib-cr. CONCLUSION: These data suggest that aquatic products are an important antibiotic resistance reservoir and highlight possible risks regarding food safety.

Enhanced heterogeneous interface to construct intelligent conductive hydrogel gas sensor for individualized treatment of infected wounds.

In this study, we developed an enhanced heterogeneous interface intelligent conductive hydrogel NH3 sensor for individualized treatment of infected wounds. The sensor achieved monitoring, self-diagnosis, and adaptive gear adjustment functions. The PPY@PDA/PANI(3/6) sensor had a minimum NH3 detection concentration of 50 ppb and a response value of 2.94 %. It also had a theoretical detection limit of 49 ppt for infected wound gas. The sensor exhibited a fast response time of 23.2 s and a recovery time of 42.9 s. Tobramycin (TOB) was encapsulated in a self-healing QCS/OD hydrogel formed by quaternized chitosan (QCS) and oxidized dextran (OD), followed by the addition of polydopamine-coated polypyrrole nanowires (PPY@PDA) and polyaniline (PANI) to prepare electrically conductive drug-loaded PPY@PDA/PANI hydrogels. The drug-loaded PPY@PDA/PANI hydrogel was combined with a PANI/PVDF membrane to form an enhanced heterogeneous interfacial PPY@PDA/PANI/PVDF-based sensor, which could adaptively learn the individual wound ammonia response and adjust the speed of drug release from the PPY@PDA/PANI hydrogel with electrical stimulation. Drug release and animal studies demonstrated the efficacy of the PPY@PDA/PANI hydrogel in inhibiting infection and accelerating wound healing. In conclusion, the gas-sensitive conductive hydrogel sensing system is expected to enable intelligent drug delivery and provide personalized treatment for complex wound management.

Advances in the Therapeutic Applications of Plant-Derived Exosomes in the Treatment of Inflammatory Diseases.

Plant-derived exosomes (PLDEs) are small extracellular vesicles that encapsulate proteins, nucleic acids and lipids, and they are usually involved in intercellular communication and molecular transport in plants. PLDEs are widely used in the therapy of diseases due to their abundance and easy availability. The diverse roles of PLDEs, which include transportation of drugs, acting as biomarkers for diagnosis of diseases and their roles in different therapies, suggest that there is a need to fully understand all the mechanisms involved in order to provide the optimum conditions for their therapeutic use. This review summarizes the biogenesis, components and functions of PLDEs and focuses on their use as therapeutic agents in the treatment of inflammatory diseases. It also explores new ideas for novel approaches in which PLDEs could potentially help patients with inflammatory diseases in the future.

Optogenetics-inspired manipulation of synaptic memory using all-optically controlled memristors.

Memristive synapses compatible with optogenetic techniques allow for the fast and low-power manipulation of memory activities using light in artificial neural systems. However, most of the optoelectronic memristors operate in the hybrid optic-electric mode; the reversible regulation of memristive states solely using light for optogenetic emulation is difficult. In this work, an all-optical controlled optoelectronic memristor (Au/Cs2AgBiBr6/Au) is developed for mimicking optogenetics-tuned memory formation and erasing behaviors in biological synapses. We show that the memristor exhibits positive and negative persistent photoconductivity effects under different light wavelengths, attributed to light-regulated carrier de-trapping/trapping at the Au/Cs2AgBiBr6 interface. This device can emulate both excitatory and inhibitory synaptic plasticity and associated learning and memory effects under light illumination. We constructed a prototype optoelectronic synaptic array and implemented the all-optically controlled memory implantation, erasing, and modification, demonstrating the light-reconfigured cognition capabilities. Our findings will inspire the development of all-optically controlled artificial neural systems with good reconfigurability for efficient neuromorphic computing and machine vision.

A new hybrid method with data-characteristic-driven analysis for artificial intelligence and robotics index return forecasting.

Forecasting returns for the Artificial Intelligence and Robotics Index is of great significance for financial market stability, and the development of the artificial intelligence industry. To provide investors with a more reliable reference in terms of artificial intelligence index investment, this paper selects the NASDAQ CTA Artificial Intelligence and Robotics (AIRO) Index as the research target, and proposes innovative hybrid methods to forecast returns by considering its multiple structural characteristics. Specifically, this paper uses the ensemble empirical mode decomposition (EEMD) method and the modified iterative cumulative sum of squares (ICSS) algorithm to decompose the index returns and identify the structural breakpoints. Furthermore, it combines the least-square support vector machine approach with the particle swarm optimization method (PSO-LSSVM) and the generalized autoregressive conditional heteroskedasticity (GARCH) type models to construct innovative hybrid forecasting methods. On the one hand, the empirical results indicate that the AIRO index returns have complex structural characteristics, and present time-varying and nonlinear characteristics with high complexity and mutability; on the other hand, the newly proposed hybrid forecasting method (i.e., the EEMD-PSO-LSSVM-ICSS-GARCH models) which considers these complex structural characteristics, can yield the optimal forecasting performance for the AIRO index returns.

On-chip Ce:YIG/Si Mach-Zehnder optical isolator with low power consumption.

The integrated optical isolator is an essential building block in photonic integrated chips. However, the performance of on-chip isolators based on the magneto-optic (MO) effect has been limited due to the magnetization requirement of permanent magnets or metal microstrips on MO materials. Here, an MZI optical isolator built on a silicon-on-insulator (SOI) without any external magnetic field is proposed. A multi-loop graphene microstrip operating as an integrated electromagnet above the waveguide, instead of the traditional metal microstrip, generates the saturated magnetic fields required for the nonreciprocal effect. Subsequently, the optical transmission can be tuned by varying the intensity of currents applied on the graphene microstrip. Compared with gold microstrip, the power consumption is reduced by 70.8%, and temperature fluctuation is reduced by 69.5% while preserving the isolation ratio of 29.44 dB and the insertion loss of 2.99 dB at1550 nm.

Near-Sensor Reservoir Computing for Gait Recognition via a Multi-Gate Electrolyte-Gated Transistor.

The recent emergence of various smart wearable electronics has furnished the rapid development of human-computer interaction, medical health monitoring technologies, etc. Unfortunately, processing redundant motion and physiological data acquired by multiple wearable sensors using conventional off-site digital computers typically result in serious latency and energy consumption problems. In this work, a multi-gate electrolyte-gated transistor (EGT)-based reservoir device for efficient multi-channel near-sensor computing is reported. The EGT, exhibiting rich short-term dynamics under voltage modulation, can implement nonlinear parallel integration of the time-series signals thus extracting the temporal features such as the synchronization state and collective frequency in the inputs. The flexible EGT integrated with pressure sensors can perform on-site gait information analysis, enabling the identification of motion behaviors and Parkinson's disease. This near-sensor reservoir computing system offers a new route for rapid analysis of the motion and physiological signals with significantly improved efficiency and will lead to robust smart flexible wearable electronics.

Synthesis and Demulsification Properties of Poly (DMDAAC-co-DAMBAC) (9:1) Copolymer.

Utilizing the copolymerization modification of dimethyl diallyl ammonium chloride (DMDAAC), the high positive charge density of the copolymer could be maintained, thereby facilitating the deficiency of its monomer in the application. In this paper, poly (DMDAAC-co-DAMBAC) (9:1) was synthesized with an aqueous polymerization method using DMDAAC and methyl benzyl diallyl ammonium chloride (DAMBAC) as monomers and 2,2'-azobis [2-methylpropionamidine] dihydrochloride (V50) as an initiator. Targeted to the product's weight-average relative molecular mass (Mw), the response surface methodology (RSM) was used to optimize the preparation process. The optimal process conditions were obtained as follows: w (M) = 80.0%, m (V50):m (M) = 0.00700%, m (Na4EDTA):m (M) = 0.00350%, T1 = 50.0 °C, T2 = 60.0 °C, and T3 = 72.5 °C. The intrinsic viscosity ([η]) of the product was 1.780 dL/g, and the corresponding double bond conversion (Conv.) was 90.25 %. Poly (DMDAAC-co-DAMBAC) (9:1) revealed a highest Mw of 5.637 × 105, together with the polydispersity index d (Mw/Mn) as 1.464. For the demulsification performance of simulated crude oil O/W emulsions, the demulsification rate of poly (DMDAAC-co-DAMBAC) (9:1) could reach 97.73%. Our study has illustrated that the copolymerization of DMDAAC and a small amount of DAMBAC with poor reactivity could significantly improve the relative molecular weight of the polymer, enhance its lipophilicity, and thus the application scope of the polymer.

A lightweight convolutional neural network hardware implementation for wearable heart rate anomaly detection.

In this article, we propose a lightweight and competitively accurate heart rhythm abnormality classification model based on classical convolutional neural networks in deep neural networks and hardware acceleration techniques to address the shortcomings of existing wearable devices for ECG detection. The proposed approach to build a high-performance ECG rhythm abnormality monitoring coprocessor achieves a high degree of data reuse in time and space, which reduces the number of data flows, provides a more efficient hardware implementation and reduces hardware resource consumption than most existing models. The designed hardware circuit relies on 16-bit floating-point numbers for data inference at the convolutional, pooling, and fully connected layers, and implements acceleration of the computational subsystem through a 21-group floating-point multiplicative-additive computational array and an adder tree. The front- and back-end design of the chip was completed on the TSMC 65 nm process. The device has an area of 0.191 mm2, a core voltage of 1 V, an operating frequency of 20 MHz, a power consumption of 1.1419 mW, and requires 5.12 kByte of storage space. The architecture was evaluated using the MIT-BIH arrhythmia database dataset, which showed a classification accuracy of 97.69% and a classification time of 0.3 ms for a single heartbeat. The hardware architecture offers high accuracy with a simple structure, low resource footprint, and the ability to operate on edge devices with relatively low hardware configurations.

Reservoir Computing-Based Design of ZnO Memristor-Type Digital Identification Circuits.

Reservoir Computing (RC) is a network architecture inspired by biological neural systems that maps time-dimensional input features to a high-dimensional space for computation. The key to hardware implementation of the RC system is whether sufficient reservoir states can be generated. In this paper, a laboratory-prepared zinc oxide (ZnO) memristor is reported and modeled. The device is found to have nonlinear dynamic responses and characteristics of simulating neurosynaptic long-term potentiation (LTP) and long-term depression (LTD). Based on this, a novel two-level RC structure based on the ZnO memristor is proposed. Novel synaptic encoding is used to maintain stress activity based on the characteristics of after-discharge and proneness to fatigue during synaptic transmission. This greatly alleviates the limitations of the self-attenuating characteristic reservoir of the duration and interval of the input signal. This makes the reservoir, in combination with a fully connected neural network, an ideal system for time series classification. The experimental results show that the recognition rate for the complete MNIST dataset is 95.08% when 35 neurons are present as hidden layers while achieving low training consumption.

Divalent cation induced re-entrant condensation behavior for lipopolysaccharides.

Lipopolysaccharides (LPSs) are negatively charged molecules covering the surface of Gram-negative bacteria (GNB). Adding divalent cations (DCs) is important to stabilize the LPS bilayer. Thus, DCs are always only considered as membrane stabilizing ions. Here, on the basis of a coarse-grained (CG) Martini force field, we conduct molecular dynamic (MD) simulations to study the divalent cation mediated LPS interaction and the stability of the LPS membrane in a wide range of DC concentrations. By measuring the LPS binding free energy and the LPS-LPS aggregate from the association course between two LPS molecules, we find that the initial addition of DCs may significantly facilitate the aggregation of LPSs into a compact structure, while sequentially adding more DCs only unpacks the LPS aggregate and drives the dissolution of LPSs. With an increasing concentration of DCs, we find a gradual replacement of DCs to monovalent cations as condensed counterions on the LPS, which follows a sign change from negative to positive in terms of the LPS effective charge and a switch of LPSs in solution from undergoing precipitation to resolubilization on adding DCs. This interaction change in the level of two LPSs accounts for the structure variation of the LPS assembly on a larger scale, where the LPS packing rigidity in the assembly bilayer is found with a similar nonmonotonic dependence with the DC concentration. Thus, our results demonstrate for the first time the presence of a re-entrant condensation behavior for LPS molecules, which can be exploited for developing novel membrane-perturbing agents based on multivalent ions as efficient GNB antibiotics.

A flexible dual-gate hetero-synaptic transistor for spatiotemporal information processing.

Artificial synapses based on electrolyte gated transistors with conductance modulation characteristics have demonstrated their great potential in emulating the memory functions in the human brain for neuromorphic computing. While previous studies are mostly focused on the emulation of the basic memory functions of homo-synapses using single-gate transistors, multi-gate transistors offer opportunities for the mimicry of more complex and advanced memory formation behaviors in biological hetero-synapses. In this work, we demonstrate an artificial hetero-synapse based on a dual-gate electrolyte transistor that can implement in situ spatiotemporal information integration and storage. We show that electric pulses applied on a single gate or unsynchronized electric pulses applied on dual gates only induce volatile conductance modulation for short-term memory emulation. In contrast, the device integrates the electric pulses coincidently applied on the dual gates in a supralinear manner and exhibits nonvolatile conductance modulation, enabling long-term memory emulation. Further studies prove that artificial neural networks based on such hetero-synaptic transistors can autonomously filter the random noise signals in the dual-gate inputs during spatiotemporal integration, facilitating the formation of accurate and stable memory. Compared to the single-gate synaptic transistor, the classification accuracy of MNIST handwritten digits using the hetero-synaptic transistor is improved from 89.3% to 99.0%. These findings demonstrate the great potential of multi-gate hetero-synaptic transistors in simulating complex spatiotemporal information processing functions and provide new platforms for the design of advanced neuromorphic computing systems.