The hottest Data science Substack posts right now

And their main takeaways
Category
Top Technology Topics

📽 Webinar: How To Maximize Model Accuracy

TheSequence • 70 implied HN points • 16 Dec 24
🕹 Technology Data science
  1. Models can lose accuracy over time in real use. It's important to know why this happens so you can fix it.
  2. Just because a model works well during training doesn't mean it will perform the same way in the real world. There are often differences that can affect results.
  3. Smart feature engineering is crucial for maintaining model accuracy without spending too much money. There are ways to improve performance that don't break the bank.

sqlmesh init duckdb

davidj.substack • 71 implied HN points • 03 Dec 24
🕹 Technology Data science
  1. There's a new public repository called bluesky-data where people can collaborate and follow along with its development. It's easy to get started by setting it up on your local machine.
  2. Using sqlmesh with the Bluesky data can provide real-time data availability, while also allowing for a more complete view of the data in a batch processing style. This means you can get both immediate updates and historical data.
  3. It's better to start with dlt and then initialize sqlmesh within that project. This way, you can efficiently manage large datasets without needing to compute everything each time.

The beginner’s guide to AI model architectures

Technically • 20 implied HN points • 05 Aug 25
🕹 Technology Data science
  1. AI models are like blueprints, guiding how models are built and designed. Choosing the right design is key to solving specific problems effectively.
  2. Neurons mimic real brain functions and are the basic units that help AI learn patterns from data. They work by performing simple math repeatedly across many layers.
  3. There are many ways to connect these neurons, forming various network types, like feedforward or recurrent networks. Each type is good for different tasks, like language or vision.

Why Making a Non-Woke AI Is Actually Very Hard

The Future of Life • 19 implied HN points • 26 Feb 24
🕹 Technology Data science
  1. Language models learn from the data they are trained on, which often includes a lot of left-leaning content, making them reflect that bias.
  2. Adjusting a model's political views is complicated because it involves changing an entire worldview, which can mess up the quality of the responses.
  3. Creating a balanced AI requires new training methods, as current models can’t easily switch perspectives without losing their effectiveness.
Get a weekly roundup of the best Substack posts, by hacker news affinity:

How I networked with one of the most prominent Data Scientists online[Storytime Saturdays]

Technology Made Simple • 59 implied HN points • 29 Jan 23
💼 Business Data science
  1. Networking is a valuable skill to add to your toolbox for personal growth, career progression, or assisting others.
  2. Even without an established online presence, you can stand out and network effectively with prominent individuals in your field.
  3. Effective networking can lead to opportunities and contracts that showcase your skills and expertise to the right people.

Synthetic Data: How to Use LLM to Improve the Performance of LLM (WizardLM)

DataSyn’s Substack • 1 HN point • 27 Aug 24
🕹 Technology Data science
  1. Synthetic data can help solve problems with real-world data, like data scarcity and privacy issues. By using artificial data, we can create large sets that are safe and more accessible.
  2. The Evol-Instruct method creates complex commands from simpler ones, which leads to richer training data for models. This process helps develop a variety of tasks for AI to learn from.
  3. Training models like WizardLM with synthetic data has shown to improve their performance significantly. It produces better responses compared to many other models, helping AI handle tougher challenges.

Catastrophic Forgetting In LLMs

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 22 Feb 24
🕹 Technology Data science
  1. Catastrophic forgetting happens when language models forget things they learned before as they learn new information. It's like a student who forgets old lessons when they study new subjects.
  2. Language models can change their performance over time, sometimes getting worse instead of better. This means they can produce different answers for the same question at different times.
  3. Continuous training can make models forget important knowledge, especially in understanding complex topics. Researchers suggest that special training techniques might help reduce this forgetting.

Tech Talks Weekly #6

Tech Talks Weekly • 19 implied HN points • 14 Mar 24
🕹 Technology Data science
  1. Tech Talks Weekly shares recent tech talks from major conferences like Devoxx and NDC. It's a great way to keep updated on the latest in tech.
  2. There's a special edition featuring over 550 talks from Kubernetes conferences. This provides a huge resource for anyone interested in cloud technology.
  3. The newsletter encourages sharing with friends and colleagues to build a community. Spreading the word helps more people connect with the tech talk content.

GroupBy #23: Meta loves Python, How Uber Serves Over 40 Million Reads Per Second from Online Storage Using an Integrated Cache

VuTrinh. • 19 implied HN points • 20 Feb 24
🕹 Technology Data science
  1. Meta is heavily invested in Python, and they're working on improvements to enhance its performance and usability.
  2. Uber has developed a powerful database called Docstore that can handle over 40 million reads per second, demonstrating their capability in data management.
  3. Data, while useful, doesn't capture the complete reality, and it's important to recognize its limitations in understanding complex scenarios.

The Tech Buffet #9: Let's talk about LLM Hallucinations

The Tech Buffet • 39 implied HN points • 24 Oct 23
🕹 Technology Data science
  1. LLMs, or Large Language Models, often produce incorrect or misleading information, known as hallucinations. This happens because they generate text based on probabilities, not actual understanding.
  2. To measure how factually accurate LLM responses are, a tool called FActScore can break down answers into simple facts and check if these facts are true. This helps in gauging the accuracy of the information given by LLMs.
  3. To reduce hallucinations, it's important to implement strategies such as allowing users to edit AI-generated content, providing citations, and encouraging detailed prompts. These methods can help improve the trustworthiness and reliability of the information LLMs produce.

ModernBERT, the BERT of 2024

Gonzo ML • 63 implied HN points • 19 Dec 24
🕹 Technology Data science
  1. ModernBERT is a new version of BERT that improves processing speed and memory efficiency. It can handle longer contexts and makes BERT more practical for today's tasks.
  2. The architecture of ModernBERT has been updated with features that enhance performance, like better attention mechanisms and optimized computations. This means it works faster and can process more data at once.
  3. ModernBERT has shown impressive results in various natural language understanding tasks and can compete well against larger models, making it an exciting tool for developers and researchers.

Friday Finds - Bonus Links

Data Science Weekly Newsletter • 19 implied HN points • 16 Feb 24
🕹 Technology Data science
  1. There are new tutorials available for those interested in AI and humanities. These tutorials aim to help people learn how to use AI tools effectively.
  2. The Leverhulme Programme is offering opportunities in ecological data science. This program is designed for doctoral training and focuses on important ecological research.
  3. A team is looking to hire a remote R programmer. They want someone to create an easy-to-use package for analyzing complex models in R.

Demonstrate, Search, Predict (DSP) for LLMs

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 16 Feb 24
🕹 Technology Data science
  1. The Demonstrate, Search, Predict (DSP) approach is a method for answering questions using large language models by breaking it down into three stages: demonstration, searching for information, and predicting an answer.
  2. This method improves efficiency by allowing for complex systems to be built using pre-trained parts and straightforward language instructions. It simplifies AI development and speeds up the creation of new systems.
  3. Decomposing queries, known as Multi-Hop or Chain-of-Thought, helps the model reason through questions step by step to arrive at accurate answers.

Data Science Weekly - Issue 479

Data Science Weekly Newsletter • 99 implied HN points • 27 Jan 23
🕹 Technology Data science
  1. Exploratory programming is important for data teams. It helps them find insights rather than just building software.
  2. Most datasets are not normally distributed, and there are many tests to check this but they can be tricky to use.
  3. AI is gaining a lot of attention, similar to what crypto once had. People are questioning if it can keep that interest alive.

The Sequence Knowledge #487: A RAG that Assesses Itself

TheSequence • 49 implied HN points • 11 Feb 25
🕹 Technology Data science
  1. Self-RAG is a new method that helps improve how retrieval-augmented generation works by letting models check their own work.
  2. It uses special tokens that help the model decide when it should look for information and how to review its own answers.
  3. This technique aims to make the process more thoughtful compared to regular methods that just pull information randomly.

sqlmesh cube_generate

davidj.substack • 59 implied HN points • 16 Dec 24
🕹 Technology Data science
  1. Building integrations can seem tough, but understanding the metadata available can simplify the process. It's important to leverage existing tools to create new functionalities efficiently.
  2. Trying out new ideas, even if they might fail, is essential for learning and discovering possibilities. Taking small steps can help you manage potential setbacks.
  3. Creating a command to generate projects based on existing data models can streamline workflows. It allows for easier implementation of complex data relationships when set up correctly.

BYOK

Aipreneur • 39 implied HN points • 08 Mar 23
🕹 Technology Data science
  1. BYOD (Bring Your Own Device) became popular in corporates due to iPhone's rise and employee preferences.
  2. BYOD is beneficial for companies in cost-saving, convenience, increased mobility, and changing workforce demographics.
  3. The emerging trend of BYOK (Bring Your Own Keys) is starting in AI platforms, where users need to pay for keys to access and use data responsibly.

Concordium and 2021.ai Team Up to Make AI More Trustworthy

Concordium Monthly Updates • 39 implied HN points • 20 Jul 23
🕹 Technology Data science
  1. Partnership between Concordium and 2021.ai enhances trust in AI through data validation and audit trails.
  2. Integration of Concordium's blockchain into 2021.ai's platform enables new use cases like ESG Validation and MiCA compliance.
  3. Collaboration aims to promote responsible and ethical use of AI, driving innovation and building trust in the AI industry.

Twitter open-sourced their recommendation algorithm

MLOps Newsletter • 39 implied HN points • 09 Apr 23
🕹 Technology Data science
  1. Twitter has open-sourced their recommendation algorithm for both training and serving layers.
  2. The algorithm involves candidate generation for in-network and out-network tweets, ranking models, and filtering based on different metrics.
  3. Twitter's recommendation algorithm is user-centric, focusing on user-to-user relationships before recommending tweets.

Introduction to R for Data Science (Part Seven Final)

The Software & Data Spectrum • 39 implied HN points • 06 Apr 23
🕹 Technology Data science
  1. Boxplots are common for visualizing data like stock pricing, and you can customize them with colors and flips.
  2. Variable plotting can include heat maps to show occurrences, and you can adjust the appearance with features like scale_fill_gradient().
  3. Coordinate your graphs using functions like coord_cartesian() and facet them based on specific variables for more detailed insights.

Introduction to R for Data Science (Part Five)

The Software & Data Spectrum • 39 implied HN points • 30 Mar 23
🕹 Technology Data science
  1. Using apply functions in R like lapply and sapply can help apply functions to elements in a vector or list.
  2. Math functions in R like abs(), sum(), mean(), and round() are useful for basic calculations and rounding numbers.
  3. Data manipulation in R using dplyr involves functions like filter(), arrange(), select(), and mutate() to filter, sort, and create new columns in datasets.

How Hugging Face and Kaggle Bolster the Open Source Machine Learning Community

The Strategy Deck • 39 implied HN points • 26 Jul 23
🕹 Technology Data science
  1. Open source ML hubs like Hugging Face and Kaggle provide platforms for managing, sharing, and deploying ML models.
  2. Hugging Face focuses on models, datasets, deployment infrastructure, and community engagement.
  3. Kaggle empowers learners, developers, and researchers with educational resources, open source models, and a competitive platform.

Python Powers Excel

Sector 6 | The Newsletter of AIM • 39 implied HN points • 24 Aug 23
🕹 Technology Data science
  1. Python is now integrated into Excel, making it easier for users to blend Excel's tools with Python's capabilities.
  2. This allows users to perform advanced tasks like data visualization and machine learning directly in Excel.
  3. The integration works well with existing Excel features, so users can still use familiar functions like formulas and charts.

For the Love of PyTorch

Sector 6 | The Newsletter of AIM • 39 implied HN points • 04 Sep 23
🕹 Technology Data science
  1. PyTorch is a key player in the development of AI, particularly large language models (LLMs). Its flexibility makes it great for deep learning experiments.
  2. The framework supports GPUs really well and allows for easy updates to computation graphs during programming.
  3. In 2022, PyTorch had a significant edge on platforms like Hugging Face, with 92% of models being PyTorch-exclusive compared to just 8% for TensorFlow.

XGBoost is the Secret of ML Energy

Sector 6 | The Newsletter of AIM • 39 implied HN points • 06 Sep 23
🕹 Technology Data science
  1. XGBoost, or Extreme Gradient Boosting, helps improve the performance and speed of machine learning models that deal with tabular data. It's known for being really good at finding patterns and making predictions.
  2. This algorithm works best for supervised learning when you have lots of training examples, especially when you have both categorical and numeric data. It can handle a mix of different data types well.
  3. If you're working with a dataset that has many features, XGBoost is a strong choice to enhance the capabilities of your machine learning model. It makes it easier to get accurate results.

Google’s Missed AI Opportunity

Sector 6 | The Newsletter of AIM • 39 implied HN points • 31 Aug 23
🕹 Technology Data science
  1. Google missed a huge chance by overlooking the Transformer paper in 2017, which cost them around $6.2 billion. This mistake allowed others to build successful AI startups.
  2. The authors of the Transformer paper have moved on to create their own companies, showing the impact of their work and how they’ve found success after leaving Google.
  3. Such missed opportunities highlight the importance of recognizing and supporting innovative research within companies like Google.

Edge 461: The Many Challenges of Kowledge Distillation

TheSequence • 56 implied HN points • 31 Dec 24
🕹 Technology Data science
  1. Knowledge distillation can be tricky because there’s a big size difference between the teacher model and the student model. The teacher model usually has a lot more parameters, making it hard to share all the useful information with the smaller student model.
  2. Transferring the complex knowledge from a large model to a smaller one isn't straightforward. The smaller model might not be able to capture all the details that the larger model has learned.
  3. Despite the benefits, there are significant challenges that need to be tackled when using knowledge distillation in machine learning. These challenges stem from the complexity and scale of the models involved.

The Counterfactual's poll #2

The Counterfactual • 19 implied HN points • 05 Feb 24
🕹 Technology Data science
  1. Subscribers can vote each month on research topics. This helps decide what the writer will explore next based on community interest.
  2. The upcoming projects mostly focus on how Large Language Models (LLMs) can measure or modify readability. Some topics might take more than a month to research thoroughly.
  3. One of the suggested studies looks at whether AI responses vary by month, testing if it seems 'lazier' in December compared to other months.

BizML: a framework for success in applied machine learning

TechTalks • 19 implied HN points • 05 Feb 24
🕹 Technology Data science
  1. Most machine learning projects fail due to a gap in understanding between data scientists and business professionals.
  2. Eric Siegel introduces bizML, a six-step framework for successful machine learning projects that emphasizes starting with the end business goal.
  3. Improving human understanding and leadership is crucial for the success of advanced technologies like machine learning.

Understanding The Role of Vector DB in AI Application

The Beep • 19 implied HN points • 04 Feb 24
🕹 Technology Data science
  1. Vector databases are designed to handle complex and unstructured data, making them great for AI applications like semantic search and face recognition. They convert information into high-dimensional vectors that are easy to work with.
  2. Unlike traditional databases, vector databases can manage different types of data such as text, images, and audio, which makes them very versatile. They're like a Swiss Army knife for managing data.
  3. Vector databases play a crucial role in enhancing AI capabilities, providing better access and analysis of data, which leads to smarter applications, including smart assistants and more.

Friday Finds - Bonus Links

Data Science Weekly Newsletter • 19 implied HN points • 02 Feb 24
🕹 Technology Data science
  1. Paid subscribers get extra links and content. It's a nice way to say thank you for their support.
  2. There are interesting discussions on topics like AI and machine learning. These conversations help people learn more about the field.
  3. Links to simulations and insights about reality powered by AI are shared. They could spark curiosity and understanding about modern technology.

The Sequence Knowledge #545 : Beyond Language, Learning About Multimodal Benchmarks

TheSequence • 28 implied HN points • 20 May 25
🕹 Technology Data science
  1. Multimodal benchmarks are tools to evaluate AI systems that use different types of data like text, images, and audio. They help ensure that AI can handle complex tasks that combine these inputs effectively.
  2. One important benchmark in this area is called MMMU, which tests AI on 11,500 questions across various subjects. This benchmark needs AI to work with text and visuals together, promoting deeper understanding rather than just shortcuts.
  3. The design of these benchmarks, like MMMU, helps reveal how well AI understands different topics and where it may struggle. This can lead to improvements in AI technology.

The Tech Buffet #23: What Nobody Tells You About RAGs

The Tech Buffet • 1 HN point • 22 Aug 24
🕹 Technology Data science
  1. It's important to understand the business needs before jumping into building a Retrieval-Augmented Generation (RAG) system. Knowing the user's context and how they will use the system will save time and improve outcomes.
  2. Different types of data need to be indexed in specific ways for a RAG to work effectively. This means treating text, images, tables, and code differently to maximize the system's performance.
  3. The quality of the data chunks you use significantly affects the answers generated by a RAG. Taking the time to create clear, relevant chunks will lead to better responses from the system.

MultiHop-RAG

Cobus Greyling on LLMs, NLU, NLP, chatbots & voicebots • 19 implied HN points • 31 Jan 24
🕹 Technology Data science
  1. Multi-hop retrieval-augmented generation (RAG) helps answer complex questions by pulling information from multiple sources. It connects different pieces of data to create a clear and complete answer.
  2. Using a data-centric approach is becoming more important for improving large language models (LLMs). This means focusing on the quality and relevance of the data to enhance how models learn and generate responses.
  3. The development of prompt pipelines in RAG systems is gaining attention. These pipelines help organize the process of retrieving and combining information, making it easier for models to handle text-related tasks.

Edge 449: Getting Into Adversarial Distillation

TheSequence • 63 implied HN points • 19 Nov 24
🕹 Technology Data science
  1. Adversarial distillation is a new model training method inspired by generative adversarial networks (GANs). It uses a setup where one part generates data and another part tries to tell if it's real or fake.
  2. This method helps improve knowledge transfer in models by combining typical distillation techniques with adversarial training. It's like guiding a student while testing their understanding.
  3. The process involves a generator that creates synthetic samples and a discriminator that distinguishes these samples from real ones, making learning more effective.