In this paper, we address the problem of compact model parameter extraction to simultaneously extract tens of parameters via derivative-free optimization. Traditionally, parameter extraction is performed manually by dividing the complete set of parameters into smaller subsets, each targeting different operational regions of the device, a process that can take several days or even weeks. Our approach streamlines this process by employing derivative-free optimization to identify a good parameter set that best fits the compact model without performing an exhaustive number of simulations. We further enhance the optimization process to address critical issues in device modeling by carefully choosing a loss function that evaluates model performance consistently across varying magnitudes by focusing on relative errors (as opposed to absolute errors), prioritizing accuracy in key operational regions of the device above a certain threshold, and reducing sensitivity to outliers. Furthermore, we utilize the concept of train-test split to assess the model fit and avoid overfitting. This is done by fitting 80% of the data and testing the model efficacy with the remaining 20%. We demonstrate the effectiveness of our methodology by successfully modeling two semiconductor devices: a diamond Schottky diode and a GaN-on-SiC HEMT, with the latter involving the ASM-HEMT DC model, which requires simultaneously extracting 35 model parameters to fit the model to the measured data. These examples demonstrate the effectiveness of our approach and showcase the practical benefits of derivative-free optimization in device modeling.

Mathematical reasoning is an important capability of large language models~(LLMs) for real-world applications. To enhance this capability, existing work either collects large-scale math-related texts for pre-training, or relies on stronger LLMs (\eg GPT-4) to synthesize massive math problems. Both types of work generally lead to large costs in training or synthesis. To reduce the cost, based on open-source available texts, we propose an efficient way that trains a small LLM for math problem synthesis, to efficiently generate sufficient high-quality pre-training data. To achieve it, we create a dataset using GPT-4 to distill its data synthesis capability into the small LLM. Concretely, we craft a set of prompts based on human education stages to guide GPT-4, to synthesize problems covering diverse math knowledge and difficulty levels. Besides, we adopt the gradient-based influence estimation method to select the most valuable math-related texts. The both are fed into GPT-4 for creating the knowledge distillation dataset to train the small LLM. We leverage it to synthesize 6 million math problems for pre-training our JiuZhang3.0 model, which only needs to invoke GPT-4 API 9.3k times and pre-train on 4.6B data. Experimental results have shown that JiuZhang3.0 achieves state-of-the-art performance on several mathematical reasoning datasets, under both natural language reasoning and tool manipulation settings. Our code and data will be publicly released in \url{https://github.com/RUCAIBox/JiuZhang3.0}.

Large language models (LLMs) often demonstrate inconsistencies with human preferences. Previous research typically gathered human preference data and then aligned the pre-trained models using reinforcement learning or instruction tuning, a.k.a. the finetuning step. In contrast, aligning frozen LLMs without requiring alignment data is more appealing. This work explores the potential of the latter setting. We discover that by integrating self-evaluation and rewind mechanisms, unaligned LLMs can directly produce responses consistent with human preferences via self-boosting. We introduce a novel inference method, Rewindable Auto-regressive INference (RAIN), that allows pre-trained LLMs to evaluate their own generation and use the evaluation results to guide rewind and generation for AI safety. Notably, RAIN operates without the need of extra data for model alignment and abstains from any training, gradient computation, or parameter updates. Experimental results evaluated by GPT-4 and humans demonstrate the effectiveness of RAIN: on the HH dataset, RAIN improves the harmlessness rate of LLaMA 30B from 82% of vanilla inference to 97%, while maintaining the helpfulness rate. On the TruthfulQA dataset, RAIN improves the truthfulness of the already-well-aligned LLaMA-2-chat 13B model by 5%.

Find open roles in Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), Computer Vision (CV), Data Engineering, Data Analytics, Big Data, and Data Science in general, filtered by job title or popular skill, toolset and products used.

Find open roles in Artificial Intelligence (AI), Machine Learning (ML), Natural Language Processing (NLP), Computer Vision (CV), Data Engineering, Data Analytics, Big Data, and Data Science in general, filtered by job title or popular skill, toolset and products used.

A China Eastern flight carrying 132 people crashed Monday. A domestic Chinese flight with 132 passengers plummeted into the mountains of southern China on Monday, likely leaving all passengers dead and investigators launching a probe into the cause. Chinese President Xi Jinping has instructed the country's emergency services to "organize a search and rescue" operation and "identify the causes" of the Boeing 737-800 crashing, according to state media. Former chairman of the National Transportation Safety Board Jim Hall told Fox News Digital on Monday that it would be "irresponsible" to speculate what caused the crash so soon after the incident, but described how the NTSB carries out investigations into major commercial crashes. This screen grab taken from video from The Paper and received via AFPTV on March 21, 2022 shows ambulances turning off onto a side road upon arrival after a China Eastern reportedly crashed in Teng County in Wuzhou City, Guangxi province.

We present the reinterpreted and reimplemented Copycat project, an architecture solving letter analogy domain problems. To support a flexible implementation change and rigor testing process, we propose a implementation method in DrRacket by using functional abstraction, naming system, initialization, and structural reference. Finally, benefits and limitations are analyzed for cognitive architectures along the lines of Copycat.

Professor Moshe BenBassat has been a leader in Artificial Intelligence for several decades. During a long academic career with positions at Tel Aviv University, USC, and UCLA, Professor BenBassat made significant contributions in Pattern Recognition, Artificial Intelligence, Optimization, Data Science and Machine Learning. Following his invention of "service chain optimization," he founded ClickSoftware which has been leading this space with AI-centric products since its inception in 1997. Moshe served as ClickSoftware's CEO until 2015, at which point it was acquired for nearly $0.5 Billion. Moshe also founded Plataine which is focused on intelligent automation for smart manufacturing, leveraging Industrial Internet of Things (IIoT) and Artificial Intelligence technologies.

It may only be one metre tall, but a R2-D2 droid has just become one of the most valuable pieces of Hollywood memorabilia ever sold. The robot, which was used in several Star Wars films, sold at auction this week for a staggering £2.13 million ($2.76 million). While those who haven't seen the Star Wars films might see this as a big investment, experts claim that the droid is the'creme de la creme of movie props' and something that you could'put next to a Picasso.' The robot, which was used in several Star Wars films, sold at auction this week for a staggering £2.13 million ($2.76 million) The R2-D2 droid was sold by Profiles in History, an auction house based in Calabasas, California, as part of its Hollywood Auction. The one metre (43 inch) tall unit was compiled from parts used throughout filming of the original Star Wars trilogy, from 1977-1983.