arXiv subjects

Alper Demir

Publications and source records attributed to Alper Demir.

13 recordsLinked to original sources

arXiv2026

Fairness over Equality: Correcting Social Incentives in Asymmetric Sequential Social Dilemmas

Sequential Social Dilemmas (SSDs) provide a key framework for studying how cooperation emerges when individual incentives conflict with collective welfare. In Multi-Agent Reinforcement Learning, these problems are often addressed by incorporating intrinsic drives that encourage prosocial or fair behavior. However, most existing methods assume that agents face identical incentives in the dilemma and require continuous access to global information about other agents to assess fairness. In this work, we introduce asymmetric variants of well-known SSD environments and examine how natural differences between agents influence cooperation dynamics. Our findings reveal that existing fairness-based methods struggle to adapt under asymmetric conditions by enforcing raw equality that wrongfully incentivize defection. To address this, we propose three modifications: (i) redefining fairness by accounting for agents' reward ranges, (ii) introducing an agent-based weighting mechanism to better handle inherent asymmetries, and (iii) localizing social feedback to make the methods effective under partial observability without requiring global information sharing. Experimental results show that in asymmetric scenarios, our method fosters faster emergence of cooperative policies compared to existing approaches, without sacrificing scalability or practicality.

cs.LG↗

arXiv2023

Self-Sustaining Oscillator with Frequency Counter for Resonance Frequency Tracking in Micro- and Nanomechanical Sensing

Nanomechanical sensors based on detecting and tracking resonance frequency shifts are to be used in many applications. Various open- and closed-loop tracking schemes, all offering a trade-off between speed and precision, have been studied both theoretically and experimentally. In this work, we advocate the use of a frequency counter as a frequency shift monitor in conjunction with a self-sustaining oscillator (SSO) nanoelectromechanical system (NEMS) configuration. We derive a theoretical model for characterizing the speed and precision of frequency measurements with state-of-the-art frequency counters. Based on the understanding provided by this model, we introduce novel enhancements to frequency counters that result in a trade-off characteristics which is on a par with the other tracking schemes. We describe a low-cost field-programmable-gate array (FPGA) based implementation for the proposed frequency counter and use it with the SSO-NEMS device in order to study its frequency tracking performance. We compare the proposed approach with the phase-locked-loop based scheme both in theory and experimentally. Our results show that similar or better performance can be achieved at a substantially lower cost and improved ease-of-use. We obtain almost perfect correspondence between the theoretical model predictions and the experimental measurements.

eess.SP↗

arXiv2023

Machine Learning and Kalman Filtering for Nanomechanical Mass Spectrometry

Nanomechanical resonant sensors are used in mass spectrometry via detection of resonance frequency jumps. There is a fundamental trade-off between detection speed and accuracy. Temporal and size resolution are limited by the resonator characteristics and noise. A Kalman filtering technique, augmented with maximum-likelihood estimation, was recently proposed as a Pareto optimal solution. We present enhancements and robust realizations for this technique, including a confidence boosted thresholding approach as well as machine learning for event detection. We describe learning techniques that are based on neural networks and boosted decision trees for temporal location and event size estimation. In the pure learning based approach that discards the Kalman filter, the raw data from the sensor are used in training a model for both location and size prediction. In the alternative approach that augments a Kalman filter, the event likelihood history is used in a binary classifier for event occurrence. Locations and sizes are predicted using maximum-likelihood, followed by a Kalman filter that continually improves the size estimate. We present detailed comparisons of the learning based schemes and the confidence boosted thresholding approach, and demonstrate robust performance for a practical realization.

physics.ins-det↗

arXiv2023

Resonance frequency tracking schemes for micro- and nanomechanical resonators

Nanomechanical resonators can serve as high performance detectors and have potential to be widely used in the industry for a variety of applications. Most nanomechanical sensing applications rely on detecting changes of resonance frequency. In commonly used frequency tracking schemes, the resonator is driven at or close to its resonance frequency. Closed-loop systems can continually check whether the resonator is at resonance and accordingly adjust the frequency of the driving signal. In this work, we study three resonance frequency tracking schemes, a feedback-free (FF), a self-sustaining oscillator (SSO), and a phase-locked loop oscillator (PLLO) scheme. We improve and extend the theoretical models for the FF and the SSO tracking schemes, and test the models experimentally with a nanoelectromechanical system (NEMS) resonator. We employ a SSO architecture with a pulsed positive feedback topology and compare it to the commonly used PLLO and FF schemes. We show that all tracking schemes are theoretically equivalent and that they all are subject to the same speed versus accuracy trade-off characteristics. In order to verify the theoretical models, we present experimental steady-state measurements for all tracking schemes. Frequency stability is characterized by computing the Allan deviation. We obtain almost perfect correspondence between the theoretical models and the experimental measurements. These results show that the choice of the tracking scheme is dictated by cost, robustness and usability in practice as opposed to fundamental theoretical differences in performance.

physics.ins-det↗

arXiv2021

Learning What to Memorize: Using Intrinsic Motivation to Form Useful Memory in Partially Observable Reinforcement Learning

Reinforcement Learning faces an important challenge in partial observable environments that has long-term dependencies. In order to learn in an ambiguous environment, an agent has to keep previous perceptions in a memory. Earlier memory based approaches use a fixed method to determine what to keep in the memory, which limits them to certain problems. In this study, we follow the idea of giving the control of the memory to the agent by allowing it to have memory-changing actions. This learning mechanism is supported by an intrinsic motivation to memorize rare observations that can help the agent to disambiguate its state in the environment. Our approach is experimented and analyzed on several partial observable tasks with long-term dependencies and compared with other memory based methods.

cs.LG↗

arXiv2021

Adaptive Time-Resolved Mass Spectrometry with Nanomechanical Resonant Sensors

Nanomechanical resonant sensors that are based on detecting and tracking the resonance frequency deviations due to events of interest are being advocated for a variety of applications. All sensor schemes currently in use are subject to a basic trade-off between accuracy and speed, while there is great interest in improving both in order to enable unprecedented and widespread applications. Based on a thorough understanding of the characteristics of current resonant sensor architectures, we propose adaptive and flexible sensor schemes. Unlike recently proposed time-resolved mechanical detection methods, the proposed schemes do not require ensemble averaging of the resonator response for many independent identical stimuli. Distinct one-time events can be detected in real-time with high time resolution with an accuracy that then improves considerably with elapsed time. While the proposed adaptive schemes also need to abide by the fundamental speed versus accuracy trade-off, we show that there is still "some room at the bottom" for improvement with sensor architecture innovations. Pareto optimal performance that reaches a bound that is imposed by the fundamental thermomechanical noise can be achieved.

physics.app-ph↗

arXiv2020

Understanding Fundamental Tradeoffs in Nanomechanical Resonant Sensors

Nanomechanical resonators are used as high performance detectors in a variety of applications such as mass spectrometry and atomic force microscopy. Initial emphasis in nanomechanical resonant sensor research was on increasing the sensitivity to the level of a single molecule, atom and beyond. On the other hand, there are applications where the speed of detection is crucial, prompting recent works that emphasize sensing schemes with improved time resolution. We first develop a general modeling framework encompassing all resonator tracking schemes currently in use, by extending recent previous work. We then explore the fundamental trade-offs between accuracy and speed in three resonant sensor architectures, namely the feedback-free open-loop approach, positive-feedback based self-sustaining oscillator, and negative-feedback based frequency-locked loop scheme. We comparatively analyze them in a unified manner, clarify some misconceptions that seem to exist in the literature, and unravel their speed versus accuracy characteristics.

physics.app-ph↗

arXiv2020

Frequency fluctuations in nanomechanical silicon nitride string resonators

High quality factor ($Q$) nanomechanical resonators have received a lot of attention for sensor applications with unprecedented sensitivity. Despite the large interest, few investigations into the frequency stability of high-$Q$ resonators have been reported. Such resonators are characterized by a linewidth significantly smaller than typically employed measurement bandwidths, which is the opposite regime to what is normally considered for sensors. Here, the frequency stability of high-$Q$ silicon nitride string resonators is investigated both in open-loop and closed-loop configurations. The stability is here characterized using the Allan deviation. For open-loop tracking, it is found that the Allan deviation gets separated into two regimes, one limited by the thermomechanical noise of the resonator and the other by the detection noise of the optical transduction system. The point of transition between the two regimes is the resonator response time, which can be shown to have a linear dependence on $Q$. Laser power fluctuations from the optical readout is found to present a fundamental limit to the frequency stability. Finally, for closed-loop measurements, the response time is shown to no longer be intrinsically limited but instead given by the bandwidth of the closed-loop tracking system. Computed Allan deviations based on theory are given as well and found to agree well with the measurements. These results are of importance for the understanding of fundamental limitations of high-$Q$ resonators and their application as high performance sensors.

cond-mat.mes-hall↗

arXiv2019

Phase-Locked Loop based Resonant Sensors: A Rigorous Theory and General Analysis Framework for Deciphering Fundamental Sensitivity Limitations due to Noise

Nanomechanical resonators are used in building ultra-sensitive mass and force sensors. In a widely used resonator based sensing paradigm, each modal resonance frequency is tracked with a phase-locked loop (PLL) based system. There is great interest in deciphering the fundamental sensitivity limitations due to inherent noise and fluctuations in PLL based resonant sensors to improve their performance. In this paper, we present a precise, first-principles based theory for the analysis of PLL based resonator tracking systems. Based on this theory, we develop a general, rigorously-derived noise analysis framework for PLL based sensors. We apply this framework to a setting where the sensor performance is mainly limited by the thermomechanical noise of the nanomechanical resonator. The results that are deduced through our analysis framework are in complete agreement with the ones we obtain from extensive, carefully run stochastic simulations of a PLL based sensor system. We compare the conclusions we derive with the recent results in the literature. Our theory and analysis framework can be used in assessing PLL based sensor performance with other sources of noise, e.g., from the electronic components, actuation and sensing mechanisms, and due to the signal generator, as well as for a variety of PLL based sensor configurations such as multi-mode and nonlinear sensing.

physics.app-ph↗

arXiv2019

Nonlinear Nanomechanical Mass Spectrometry at the Single-Nanoparticle Level

Nanoelectromechanical Systems (NEMS) have emerged as a promising technology for performing the mass spectrometry of large biomolecules and nanoparticles. As nanoscale objects land on NEMS sensor one by one, they induce resolvable shifts in the resonance frequency of the sensor proportional to their weight. The operational regime of NEMS sensors is often limited by the onset-of-nonlinearity, beyond which the highly sensitive schemes based on frequency tracking by phase-locked loops cannot be readily used. Here, we develop a measurement architecture to operate at the nonlinear regime and measure frequency shifts induced by analytes in a rapid and sensitive manner. We used this architecture to individually characterize the mass of gold nanoparticles and verified the results by performing independent measurements of the same nanoparticles based on linear mass sensing. Once the feasibility of the technique is established, we have obtained the mass spectrum of a 20 nm gold nanoparticle sample by individually recording about five hundred single particle events using two modes working sequentially in the nonlinear regime. The technique obtained here can be used for thin nanomechanical structures which possess a limited dynamic range.

cond-mat.mes-hall↗

arXiv2012

DSP Based PMD Emulators for Built-in Testing of Coherent Optical Receivers

We propose discrete-time polarization mode dispersion (PMD) models that are compatible with the emerging coherent receiver techniques, and statistical sampling schemes for the model parameters. These models use multiple-input multiple-output (MIMO) finite impulse response (FIR) filters that are lossless and therefore lend themselves as perfect candidates for emulation of fiber channels suffering from PMD without polarization dependent loss (PDL). The concatenated composition of these filters resembles the continuous time lumped model of PMD channels and offers a flexible emulator and compensator structure in terms of computational complexity which constitutes the main bottleneck for real-time DSP applications.

physics.optics↗

arXiv2012

Random Input Sampling for Complex Models Using Markov Chain Monte Carlo

Many random processes can be simulated as the output of a deterministic model accepting random inputs. Such a model usually describes a complex mathematical or physical stochastic system and the randomness is introduced in the input variables of the model. When the statistics of the output event are known, these input variables have to be chosen in a specific way for the output to have the prescribed statistics. Because the probability distribution of the input random variables is not directly known but dictated implicitly by the statistics of the output random variables, this problem is usually intractable for classical sampling methods. Based on Markov Chain Monte Carlo we propose a novel method to sample random inputs to such models by introducing a modification to the standard Metropolis-Hastings algorithm. As an example we consider a system described by a stochastic differential equation (sde) and demonstrate how sample paths of a random process satisfying this sde can be generated with our technique.

stat.ML↗

arXiv2008

Fast Monte Carlo Estimation of Timing Yield: Importance Sampling with Stochastic Logical Effort (ISLE)

In the nano era in integrated circuit fabrication technologies, the performance variability due to statistical process and circuit parameter variations is becoming more and more significant. Considerable effort has been expended in the EDA community during the past several years in trying to cope with the so-called statistical timing problem. Most of this effort has been aimed at generalizing the static timing analyzers to the statistical case. In this paper, we take a pragmatic approach in pursuit of making the Monte Carlo method for timing yield estimation practically feasible. The Monte Carlo method is widely used as a golden reference in assessing the accuracy of other timing yield estimation techniques. However, it is generally believed that it can not be used in practice for estimating timing yield as it requires too many costly full circuit simulations for acceptable accuracy. In this paper, we present a novel approach to constructing an improvedMonte Carlo estimator for timing yield which provides the same accuracy as the standard Monte Carlo estimator, but at a cost of much fewer full circuit simulations. This improved estimator is based on a novel combination of a variance reduction technique, importance sampling, and a stochastic generalization of the logical effort formalism for cheap but approximate delay estimation. The results we present demonstrate that our improved yield estimator achieves the same accuracy as the standard Monte Carlo estimator at a cost reduction reaching several orders of magnitude.

cs.OH↗