Parameter-efficient fine-tuning of large-scale pre-trained language models

Efficient and Robust Parameter Tuning for Heuristic Algorithms

The main advantage of heuristic or metaheuristic algorithms compared to exact optimization methods is their ability in handling large-scale instances within a reasonable time, albeit at the expense of losing a guarantee for achieving the optimal solution. Therefore, metaheuristic techniques are appropriate choices for solving NP-hard problems to near optimality. Since the parameters of heuristi...

Large scale structure and supersymmetric inflation without fine tuning.

We explore constraints on the spectral index n of density fluctuations and the neutrino energy density fraction ΩHDM , employing data from a variety of large scale observations. The best fits occur for n ≈ 1 and ΩHDM ≈ 0.15 − 0.30, over a range of Hubble constants 40 − 60 km s−1 Mpc−1. We present a new class of inflationary models based on realistic supersymmetric grand unified theories which d...

Efficient Model Synchronization of Large-Scale Models

COMPUTATIONALLY EFFICIENT OPTIMUM DESIGN OF LARGE SCALE STEEL FRAMES

Computational cost of metaheuristic based optimum design algorithms grows excessively with structure size. This results in computational inefficiency of modern metaheuristic algorithms in tackling optimum design problems of large scale structural systems. This paper attempts to provide a computationally efficient optimization tool for optimum design of large scale steel frame structures to AISC...

Large Scale Hierarchical Neural Network Language Models

Feed-forward neural network language models (NNLMs) are known to improve both perplexity and word error rate performance for speech recognition compared with conventional ngram language models. We present experimental results showing how much the WER can be improved by increasing the scale of the NNLM, in terms of model size and training data. However, training time can become very long. We imp...