Amit Lal

Shubham Jadhav, Kaustav Roy, Luis Amaro et al.

Neuromorphic computing demands synaptic elements that can store and update weights with high precision while being read non-destructively. Conventional ferroelectric synapses store weights in remnant polarization states and might require destructive electrical readout, limiting endurance and reliability. We demonstrate a ferroelectric MEMS (FeMEMS) based synapse in which analog weights are stored in the piezoelectric coefficient $d_{31,eff}$ of a released Hf$_{0.5}$Zr$_{0.5}$O$_2$ (HZO) MEMS unimorph. Partial switching of ferroelectric domains modulates $d_{31,eff}$, and a low-amplitude mechanical drive reads out the weight without read-disturb in the device yielding more than 7-bit of programming levels. The mechanical switching distribution function follows a Lorentzian distribution as a logarithmic function of partial poling voltage ($V_p$) consistent with nucleation-limited switching (NLS), and the median threshold extracted from electromechanical data obeys a Merz-type field-time law with a dimensionless exponent $α= 3.62$. These relationships establish a quantitative link between mechanical weights and electrical switching kinetics. This mechanically read synapse avoids depolarization and charge-injection effects, provides bipolar weights (well suited for excitatory and inhibitory synapses), directly reveals partial domain populations, and offers a robust, energy-efficient route toward high-bit neuromorphic hardware.

Amit Lal, Chunyan Shan, Peng Xi

Structured illumination microscopy (SIM) is a very important super-resolution microscopy technique, which provides high speed super-resolution with about two-fold spatial resolution enhancement. Several attempts aimed at improving the performance of SIM reconstruction algorithm have been reported. However, most of these highlight only one specific aspect of the SIM reconstruction -- such as the determination of the illumination pattern phase shift accurately -- whereas other key elements -- such as determination of modulation factor, estimation of object power spectrum, Wiener filtering frequency components with inclusion of object power spectrum information, translocating and the merging of the overlapping frequency components -- are usually glossed over superficially. In addition, most of the work reported lie scattered throughout the literature and a comprehensive review of the theoretical background is found lacking. The purpose of the present work is two-fold: 1) to collect the essential theoretical details of SIM algorithm at one place, thereby making them readily accessible to readers for the first time; and 2) to provide an open source SIM reconstruction code (named OpenSIM), which enables users to interactively vary the code parameters and study it's effect on reconstructed SIM image.

Amit Lal

Tabular foundation models aim to make machine learning accessible for enterprise data without task-specific training. This paper presents the first independent evaluation of SAP's Retrieval Pretrained Transformer (RPT-1) from a practitioner perspective. RPT-1 is a compact 64.6 MB model pretrained on 1.34 TB of structured data across 3.1 million tables. We benchmark it against tuned gradient-boosted decision trees (XGBoost, LightGBM, CatBoost) on three SAP business scenarios: demand forecasting across SD/MM/PP modules, predictive data integrity in BC/MM/QM, and financial risk classification in FI/CO/AR. Across five-fold cross-validation on datasets ranging from 2,500 to 3,200 rows, RPT-1 reaches 91-96% of tuned GBDT accuracy without any training examples. The classification gap is modest at 3.6-4.1 percentage points on AUC-ROC, though regression tasks show wider gaps of 8.9-11.1 percentage points on R-squared. An interesting finding is a crossover at roughly 75-100 context rows where RPT-1 actually outperforms XGBoost under limited data. Based on these results, we propose a practical hybrid workflow: use RPT-1 for rapid screening, then train GBDT selectively where prediction accuracy justifies the effort. All experiments are reproducible through publicly available Hugging Face Spaces.