#deep-learning

[ follow ]
fromMedium
1 week ago

How Neural Networks Are Improving Drug Discovery

Neural networks are some of the most promising artificial intelligence (AI) models. These systems process data similarly to the human brain, passing information through a complex network of nodes in the same way information goes through layers of neurons. That makes them capable of solving complicated problems in minimal time, which is particularly advantageous in the medical industry. Drug discovery is a vital but challenging process.
Artificial intelligence
fromFortune
2 weeks ago

AI safety tip: if you don't want it giving bioweapon instructions, maybe don't put them in the training data, say researchers

Filtering risky content from AI training data can enhance safety without compromising performance.
#keras
fromRlancemartin
3 weeks ago

Learning the Bitter Lesson

"The Bitter Lesson has been learned over and over across many domains of AI research, such as Chess, Go, speech, vision."
Healthcare
fromMedium
1 month ago

AI Model Detects Hidden Heart Disease from ECGs, Outperforming Cardiologists

EchoNext is an AI model that detects structural heart disease from standard 12-lead ECGs, outperforming cardiologists in diagnostic accuracy.
#tensorflow
fromHackernoon
3 months ago

Implementing Input Masking and Padding in TensorFlow Keras Models | HackerNoon

Masking informs sequence-processing layers to skip specific timesteps in input data.
fromHackernoon
2 years ago

Using LLVM To Supercharge AI Model Execution On Edge Devices | HackerNoon

LLVM simplifies optimizing AI workloads for edge devices, transforming deployment pipelines into efficient processes.
fromHackernoon
1 year ago

How Researchers Are Preprocessing Gigabyte-Sized WSIs for Deep Learning | HackerNoon

Whole slide images (WSIs) can be too large to be processed directly by Deep Learning models due to their sizes, often exceeding 6 GB of memory.
Artificial intelligence
#advertising
Cancer
fromNews Center
1 month ago

AI Matches Doctors in Mapping Lung Tumors for Radiation Therapy - News Center

iSeg, a new AI tool, enhances tumor segmentation in radiation therapy for improved cancer treatment precision and consistency.
Artificial intelligence
fromHackernoon
1 month ago

Keep the Channel, Change the Filter: A Smarter Way to Fine-Tune AI Models | HackerNoon

Decomposing convolution filters into filter atoms and coefficients enables new model tuning methods without generating intermediate features.
Digital life
fromRealpython
1 month ago

Use TorchAudio to Prepare Audio Data for Deep Learning Quiz - Real Python

The quiz tests knowledge on audio fundamentals and using TorchAudio for deep learning preparation.
#ai
fromInfoQ
2 months ago
Artificial intelligence

Nvidia's GB200 NVL72 Supercomputer Achieves 2.7 Faster Inference on DeepSeek V2

fromInfoQ
2 months ago
Artificial intelligence

Agentica Project's Open Source DeepCoder Model Outperforms OpenAI's O1 on Coding Benchmarks

fromScienceDaily
4 months ago
Artificial intelligence

Brain-inspired AI breakthrough: Making computers see more like humans

The Lp-Convolution AI technique enhances image recognition efficiency by dynamically adjusting filter shapes, mirroring human brain functions.
fromSocial Media Today
4 months ago
Vue

Google Launches Expanded AI Text-to-Video Generation Tools

Gemini Advanced users can now create high-quality 8-second videos from simple text prompts using the Veo 2 AI model.
fromInfoQ
2 months ago
Artificial intelligence

Nvidia's GB200 NVL72 Supercomputer Achieves 2.7 Faster Inference on DeepSeek V2

fromInfoQ
2 months ago
Artificial intelligence

Agentica Project's Open Source DeepCoder Model Outperforms OpenAI's O1 on Coding Benchmarks

Artificial intelligence
fromScienceDaily
4 months ago

Brain-inspired AI breakthrough: Making computers see more like humans

The Lp-Convolution AI technique enhances image recognition efficiency by dynamically adjusting filter shapes, mirroring human brain functions.
Vue
fromSocial Media Today
4 months ago

Google Launches Expanded AI Text-to-Video Generation Tools

Gemini Advanced users can now create high-quality 8-second videos from simple text prompts using the Veo 2 AI model.
#artificial-intelligence
fromWIRED
2 months ago
Artificial intelligence

How to Make AI Faster and Smarter-With a Little Help from Physics

Artificial intelligence
fromMedium
3 months ago

The Future of AI: Building World Models, Agentic Systems, and Biological Simulations

AI is transitioning from task execution to understanding and adaptation through world models, agentic AI, and AI-driven biology.
fromThe Hacker News
4 months ago
Artificial intelligence

Artificial Intelligence - What's all the fuss?

AI simulates human intelligence in machines, allowing for advanced problem-solving and decision-making.
Artificial intelligence
fromWIRED
2 months ago

How to Make AI Faster and Smarter-With a Little Help from Physics

Rose Yu integrates physics into deep learning to enhance predictions and simulations for various applications including traffic and climate.
Artificial intelligence
fromMedium
3 months ago

The Future of AI: Building World Models, Agentic Systems, and Biological Simulations

AI is transitioning from task execution to understanding and adaptation through world models, agentic AI, and AI-driven biology.
fromHackernoon
2 years ago

Challenges in Web-Scale Information Retrieval: From Keywords to Embeddings | HackerNoon

Traditional keyword matching in information retrieval fails to understand user intent, which leads to irrelevant results and limits the diversity of responses, requiring query alterations to be effective.
Web development
fromArs Technica
2 months ago

How a grad student got LHC data to play nice with quantum interference

A young researcher has developed a machine learning method that significantly enhances data analysis from the Large Hadron Collider.
fromHackernoon
9 months ago

Related Work: vAttention in LLM Inference Optimization Landscape | HackerNoon

Efficient optimization of LLM inference is essential for reducing latency and improving performance in AI applications.
Scala
fromHackernoon
55 years ago

How an 8B Open Model Sets New Standards for Safe and Efficient Vision-Language AI | HackerNoon

Idefics2 emerges as a state-of-the-art vision-language model, showcasing efficiency and performance improvements through systematic experimentation.
Scala
fromHackernoon
2 months ago

New Secure Computing Tech Handles AI, Video, and Data Without Leaking Secrets | HackerNoon

The study evaluates a framework for secure collaboration focusing on deep learning applications and privacy preservation during training and inference.
Bootstrapping
fromHackernoon
2 months ago

Build Smarter Models with Keras Functional API | HackerNoon

The functional API facilitates the use of shared layers, enabling efficient model training by reusing layer instances.
fromHackernoon
2 months ago

Smarter Satellite Vision with Few-Shot Learning | HackerNoon

Few-shot learning significantly enhances remote sensing tasks like object detection and segmentation by minimizing reliance on extensive training data.
fromHackernoon
2 months ago

How Few-Shot Learning Enhances UAV-Based Image Analysis | HackerNoon

Few-shot learning enhances deep learning applicability in remote sensing by enabling effective learning with limited labeled data.
Artificial intelligence
fromTechCrunch
2 months ago

The best ways to build on top of foundation models, with DeepMind, Twelve Labs, and Amazon | TechCrunch

Startups can capitalize on evolving AI models to build innovative and sustainable businesses.
fromHackernoon
4 years ago

Teaching Old Preconditioners New Tricks: How GNNs Supercharge Linear Solvers | HackerNoon

Recent advancements suggest that employing small Graph Neural Networks to craft preconditioners could significantly enhance performance while preserving necessary sparsity, thereby optimizing computational efficacy.
Artificial intelligence
fromHackernoon
4 months ago

LightCap's Success on Nocaps: Limitations and Opportunities for Growth | HackerNoon

The proposed framework exhibits super-balanced performance and efficiency, but has limitations such as the computational cost of the visual backbone and restricted training data.
Data science
fromClickUp
3 months ago

Glean vs. Perplexity AI: Which is Best for Knowledge Management?

Glean is an AI-powered enterprise search and knowledge management platform that helps employees quickly find the information they need, no matter where it's stored.
Artificial intelligence
Artificial intelligence
fromPsychology Today
3 months ago

AI Predicts Active Brain Cell Types With High Accuracy

AI deep learning algorithms can accurately distinguish between different brain cell types from activity recordings, marking a significant advancement in neuroscience.
fromHackernoon
4 months ago

Decoding Diffusion Models: Core Concepts & PyTorch Code | HackerNoon

Diffusion models generate high-quality data by gradually adding noise and then learning to reverse the process.
fromHackernoon
11 months ago

Even AI Needs Glasses: When Space Images Get Too Fuzzy to Fix | HackerNoon

Transformers enhance astronomical image restoration but struggle with high noise levels.
Artificial intelligence
fromInfoWorld
4 months ago

Vector Institute aims to clear up confusion about AI model performance

DeepSeek and OpenAI's o1 models excel in performance, yet AI models still face significant challenges across various tasks.
Artificial intelligence
fromDevOps.com
4 months ago

DeepCoder: Revolutionizing Software Development with Open-Source AI - DevOps.com

DeepCoder-14B-Preview is a revolutionary open-source model that democratizes AI coding tools and competes with proprietary models.
[ Load more ]