Viral News - CybAI news

23 May

Hugging Face Autotrain – Getting Started

stp2y0 CommentsViral News

Autotrain is a no-code platform from Hugging Face to train, evaluate, and deploy machine learning and deep learning models. In this article, we will use Hugging Face Autotrain to train a Small Language Model (SLM). The Hugging Face Autotrain platform offers several functionalities for training: Computer Vision models Machine Learning models And LMs & LLMs However, in this article we will focus on training a language model for instruction following using the Autotrain platform. Although we can directly access Autotrain from their platform, we will use local installation. So, in way, we will use some code rather than the no-code…

23 May

Self-Driving Cars vs. Coding Copilots

stp2y0 CommentsViral News

Back in the mid-2010s, the world of autonomous vehicles was making great progress, and it seemed that we would soon be ushered around in cars that drove themselves, leaving us free to spend our time how we wanted. That obviously hasn’t happened, but instead, we’ve been treated to a form of AI we weren’t expecting: generative AI-powered copilots. Following the launch of ChatGPT in late 2022, the world of generative AI has been on a tear. Every company seems to be investing in large language models (LLMs) to build one of the two most visible forms of GenAI: chatbots and…

23 May

Data Machina #252

stp2y0 CommentsViral News

Diffusion, FM & Pre-Trained AI models for Time-Series. DeepNN-based models are starting to match or even outperform statistical time-series analysis & forecasting methods in some scenarios. Yet, DeepNN-based models for time-series suffer from 4 key issues: 1) complex architecture 2) enormous amount of time required for training 3) high inference costs, and 4) poor context sensitivity.Latest innovative approaches. To address those issues, a new breed of foundation or pre-trained AI models for time-series is emerging. Some of these new AI models use hybrid approaches borrowing from NLP, vision/ image, or physics modelling, like: transformers, diffusion models, KANs and state space…

23 May

The Role of Synthetic Data in Cybersecurity

stp2y0 CommentsViral News

Data's value is something of a double-edged sword. On one hand, digital data lays the groundwork for powerful AI applications, many of which could change the world for the better. Conversely, storing so many details on people creates huge privacy risks. Synthetic data provides a possible solution. What Is Synthetic Data? Synthetic data is a subset of anonymized data – data that doesn't reveal any real-world details. More specifically, it refers to information that looks and acts like real-world data but has no ties to actual people, places or events. In short, it's fake data that can produce real results. In…

22 May

Announcing General Availability of Liquid Clustering

stp2y0 CommentsViral Newsdata management, delta lake, liquid clustering

We’re excited to announce the General Availability of Delta Lake Liquid Clustering in the Databricks Data Intelligence Platform. Liquid Clustering is an innovative data management technique that replaces table partitioning and ZORDER so you no longer have to fine-tune your data layout to achieve optimal query performance. Liquid clustering significantly simplifies data layout-related decisions and provides the flexibility to redefine clustering keys without data rewrites. It allows data layout to evolve alongside analytic needs over time – something you could never do with partitioning on Delta. Since the Public Preview of Liquid Clustering at the Data and AI Summit last year, we’ve…

22 May

Top Data Validation Tools for Machine Learning in 2024

stp2y0 CommentsViral News

Image generated with MidjourneyIt was challenging to stop myself from starting this article with some variation of the popular phrase "garbage in, garbage out." Well, I did it anyway. But jokes aside, we can easily imagine a situation in which we have built and deployed a machine learning model (possibly a black box) that accepts some input and returns some predictions. So far, so good.However, with tons of complexity happening before the model (data preprocessing and manipulation), the model itself, and any post-processing of the outputs, many things can go wrong. And in some mission-critical fields (finance, healthcare, or security),…

22 May

Fast Stochastic Policy Gradient: Negative Momentum for Reinforcement Learning

stp2y0 CommentsViral News

arXiv:2405.12228v1 Announce Type: new Abstract: Stochastic optimization algorithms, particularly stochastic policy gradient (SPG), report significant success in reinforcement learning (RL). Nevertheless, up to now, that how to speedily acquire an optimal solution for RL is still a challenge. To tackle this issue, this work develops a fast SPG algorithm from the perspective of utilizing a momentum, coined SPG-NM. Specifically, in SPG-NM, a novel type of the negative momentum (NM) technique is applied into the classical SPG algorithm. Different from the existing NM techniques, we have adopted a few hyper-parameters in our SPG-NM algorithm. Moreover, the computational complexity is nearly same…

22 May

Focus on Low-Resolution Information: Multi-Granular Information-Lossless Model for Low-Resolution Human Pose Estimation

stp2y0 CommentsViral News

arXiv:2405.12247v1 Announce Type: new Abstract: In real-world applications of human pose estimation, low-resolution input images are frequently encountered when the performance of the image acquisition equipment is limited or the shooting distance is too far. However, existing state-of-the-art models for human pose estimation perform poorly on low-resolution images. One key reason is the presence of downsampling layers in these models, e.g., strided convolutions and pooling layers. It further reduces the already insufficient image information. Another key reason is that the body skeleton and human kinematic information are not fully utilized. In this work, we propose a Multi-Granular Information-Lossless (MGIL) model…

22 May

The Arabic Noun System Generation

stp2y0 CommentsViral News

arXiv:2405.11014v1 Announce Type: new Abstract: In this paper, we show that the multiple-stem approach to nouns with a broken plural pattern allows for greater generalizations to be stated in the morphological system. Such an approach dispenses with truncating/deleting rules and other complex rules that are required to account for the highly allomorphic broken plural system. The generation of inflected sound nouns necessitates a pre-specification of the affixes denoting the sound plural masculine and the sound plural feminine, namely uwna and aAt, in the lexicon. The first subsection of section one provides an evaluation of some of the previous analyses of…

22 May

Fine Tuning Phi 1.5 using QLoRA on the Stanford Alpaca Dataset

stp2y0 CommentsViral News

Quantized LoRA, more commonly known as QLoRA is a combination of quantization and Low Rank Adaptation for fine-tuning LLMs. Simply put, LoRa is a technique to adapt Large Language Models to specific tasks without making them forget their pretraining knowledge. In QLoRa, we load the pretrained model weights in quantized format, say 4-bit (INT4). However, the adapter (LoRA) layers are loaded in full precision, FP16 or FP32. This reduces the memory (GPU) consumption by a great extent making fine tuning possible on low resource hardware. To this end, in this article, we will be fine tuning the Phi 1.5 model…