Stay Updated with the Latest Tech News
Get ahead of the curve with the latest insights, trends, and analysis in the tech world.
Browse by Category
Popular Topics
#General #Development #Mobile Development #Cloud & DevOps #Cybersecurity & Privacy #AI & Data Science #Hardware & Gadgets #Design & UXHow (and Where) ML Beginners Can Find Papers
From conferences to surveysContinue reading on Towards Data Science »
Share on Facebook
Share in Tweet (rip twitter)
Share on LinkedIn
Share on Reddit
Share on Threads
Share on Bluesky
What Every Aspiring Machine Learning Engineer Must Know to Succeed
Your Guide to Avoiding Critical Errors with Machine Learning in ProductionContinue reading on Towards Data Science »
Propensity-Score Matching Is the Bedrock of Causal Inference
And how to get started with it using PythonContinue reading on Towards Data Science »
Should you switch from VSCode to Cursor?
My experience using VSCode (GitHub Copilot) and Cursor (Claude 3.5 Sonnet) as a Data Scientist.Continue reading on Towards Data Science »
Evaluation-Driven Development for agentic applications using PydanticAI
An open-source, model-agnostic agentic framework that supports dependency injectionIdeally, you can evaluate agentic applications even as you are developing them, instead of evaluation being an afterthought. For this to work, though, you need to be able to mock both internal and external dependencies of the agent you are developing. I am extremely excited by PydanticAI because it supports dependency injection from the...
Conditional Variational Autoencoders for Text to Image Generation
Investigating an early generative architecture and applying it to image generation from textinputRecently I was tasked with text-to-image synthesis using a conditional variational autoencoder (CVAE). Being one of the earlier generative structures, it has its limitations but is easily implementable. This article will cover CVAEs at a high level, but the reader is presumed to have a high level understanding to cover the...
The 80/20 problem of generative AI — a UX research insight
Image byauthorThe 80/20 problem of generative AI—a UX researchinsightWhen an LLM solves a task 80% correctly, that often only amounts to 20% of the uservalue.The Pareto principle says if you solve a problem 20% through, you get 80% of the value. The opposite seems to be true for generative AI.About the author: Zsombor Varnagy-Toth is a Sr UX Researcher at SAP with background in machine learning and cognitive science....
Top 3 Strategies to Search Your Data
Strategies from traditional index seek to AI based semantic search that every software engineer should know!Continue reading on Towards Data Science »
A New Approach to AI Safety: Layer Enhanced Classification (LEC)
LEC surpasses best in class models, like GPT-4o, by combining the efficiency of a ML classifier with the language understanding of anLLMImagine sitting in a boardroom, discussing the most transformative technology of our time—artificial intelligence—and realizing we’re riding a rocket with no reliable safety belt. The Bletchley Declaration, unveiled during the AI Safety Summit hosted by the UK government and backed by...
Semantically Compress Text to Save On LLM Costs
LLMs are great… if they can fit all of yourdataPhoto by Christopher Burns onUnsplashOriginally published at https://blog.developer.bazaarvoice.com on October 28,2024.IntroductionLarge language models are fantastic tools for unstructured text, but what if your text doesn’t fit in the context window? Bazaarvoice faced exactly this challenge when building our AI Review Summaries feature: millions of user reviews simply...
When Averages Lie: Moving Beyond Single-Point Predictions
The Case for Predicting Full Probability Distributions in Decision-MakingSome people like hot coffee, some people like iced coffee, but no one likes lukewarm coffee. Yet, a simple model trained on coffee temperatures might predict that the next coffee served should be… lukewarm. This illustrates a fundamental problem in predictive modeling: focusing on single point estimates (e.g., averages) can lead us to meaningless...
Ranking Basics: Pointwise, Pairwise, Listwise
Because thy neighbour mattersImage taken from unsplash.comFirst, let’s talk about where ranking comes into play. Ranking is a big deal in e-commerce and search applications—essentially, any scenario where you need to organize documents based on a query. It’s a little different from classic classification or regression problems. For instance, in the Titanic dataset, you predict whether a passenger survives or not, and...
HuggingChat is the One LLM App You Shouldn’t Overlook
Check out this excellent chat tool you may not know about.
Understanding Deduplication Methods: Ways to Preserve the Integrity of Your Data
Increasing growth and data complexities have made data deduplication even morerelevantData duplication is still a problem for many organisations. Although data processing and storage systems have developed rapidly along with technological advances, the complexity of the data produced is also increasing. Moreover, with the proliferation of Big Data and the utilisation of cloud-based applications, today’s organisations...
How to Get Addicted to Machine Learning
A simple guide for getting hooked to machine learning and building a successful career in the field.
Introducing Layer Enhanced Classification (LEC)
A novel approach for lightweight safety classification using pruned languagemodelsLeveraging the hidden state from an intermediate Transformer layer for efficient and robust content safety and prompt injection classificationImage by author and GPT-4o meant to represent the robust language understanding provided by Large LanguageModels.IntroductionAs the adoption of Language Models (LMs) grows, it’s more and more...
5 Tools for Visualizing Machine Learning Models
Machine learning (ML) models are built upon data.
How to Stand Out as a Junior Data Scientist
7 things you can do to show your skills even if you have no experience at allContinue reading on Towards Data Science »
From Prototype to Production: Enhancing LLM Accuracy
Implementing evaluation frameworks to optimize accuracy in real-world applicationsImage created by DALL-E3Building a prototype for an LLM application is surprisingly straightforward. You can often create a functional first version within just a few hours. This initial prototype will likely provide results that look legitimate and be a good tool to demonstrate your approach. However, this is usually not enough for...
Synthetic Control Sample for Before and After A/B Test
Learn a simple way to use linear regression to create a synthetic control sample for your A/B testContinue reading on Towards Data Science »
Transform Customer Feedback into Actionable Insights with CrewAI and Streamlit
Build an AI-powered app to analyze unstructured feedback, generate insightful reports, and create interactive visualizationsContinue reading on Towards Data Science »
How to Use Docker for Local Development Environments
Learn how to create containers and manage complex setups with Docker Compose to simplify your development workflow.
2024 Highlights: The AI and Data Science Articles That Made a Splash
Feeling inspired to write your first TDS post before the end of 2024? We’re always open to contributions from newauthors.And just like that, 2024 is (almost) in the books. It was a year of exciting transitions—both for the TDS team and, in many meaningful ways, for the data science, machine learning, and AI communities at large. We’d like to thank all of you—readers, authors, and followers—for your support, and for...
A Comprehensive Guided Project to A/B Testing (+ Notebook)
From failing statistics in college to passion for hypothesis testingContinue reading on Towards Data Science »
Awesome Plotly with code series (Part 6): Dealing with long axis labels
To rotate or not rotate? To truncate or to not truncate?Continue reading on Towards Data Science »
6 Ways to Improve Your Predictive Models in Data Science
Whether you aim for building the perfect image classifier, sales predictor, or price estimator, these six pracitcal tips and insights will help you get there!
Data Science Salary Breakdown 2024
Glassdoor vs ZipRecruiter vs PayScale: U.S. comparisons to 2022Continue reading on Towards Data Science »
Introduction to TensorFlow’s Functional API
Learn what the Functional API is, and how to build complex keras models using itContinue reading on Towards Data Science »
Classifier-free guidance in LLMs Safety — NeurIPS 2024 Challenge experience
Classifier-Free Guidance in LLMs Safety—NeurIPS 2024 Challenge ExperienceThis article briefly describes NeurIPS 2024 LLM-PC submission that was awarded the second prize—the approach to effective LLM unlearning without any retaining dataset. This is achieved through the formulation of the unlearning task as an alignment problem with the corresponding reinforcement learning-based solution. The unlearning without model...
Classifier-Free Guidance in LLMs Safety — NeurIPS 2024 Challenge Experience
Classifier-Free Guidance in LLMs Safety—NeurIPS 2024 Challenge ExperienceThis article briefly describes NeurIPS 2024 LLM-PC submission that was awarded the second prize—the approach to effective LLM unlearning without any retaining dataset. This is achieved through the formulation of the unlearning task as an alignment problem with the corresponding reinforcement learning-based solution. The unlearning without model...
The Algorithm That Made Google Google
How PageRank transformed how we searched the internet, and why it’s still playing an important role in LLMs with Graph RAG.Continue reading on Towards Data Science »
100 Years of (eXplainable) AI
Reflecting on advances and challenges in deep learning and explainability in the ever-evolving era of LLMs and AI governanceImage byauthorBackgroundImagine you are navigating a self-driving car, relying entirely on its onboard computer to make split-second decisions. It detects objects, identifies pedestrians, and even can anticipate behavior of other vehicles on the road. But here’s the catch: you know it works, of...
Navigating Soft Actor-Critic Reinforcement Learning
Understanding the theory and implementation of SAC RL in the context of BioengineeringImage generated by the author using ChatGPT-4oIntroductionThe research domain of Reinforcement Learning (RL) has evolved greatly over the past years. The use of deep reinforcement learning methods such as Proximal Policy Optimisation (PPO) (Schulman, 2017) and Deep Deterministic Policy Gradient (DDPG) (Lillicrap, 2015) have enabled...
How to Write Clean Python Code as a Beginner
Writing Python code that’s clean and easy to understand isn’t just for experts — learn how to avoid common pitfalls and write like a pro from the start!
Bessel’s Correction: Why Do We Divide by n−1 Instead of n in Sample Variance?
Understanding the Unbiased Estimation of Population VarianceIn statistics, a common point of confusion for many learners is why we divide by n1 when calculating sample variance, rather than simply using n, the number of observations in the sample. This choice may seem small but is a critical adjustment that corrects for a natural bias that occurs when we estimate the variance of a population from a sample. Let’s walk...