time-series feature engineering with python itertools
build efficient time series features like lags, rolling windows, and seasonal interactions using python's itertools module.
topic
machine learning
ai agents reshape data science in 2026
ai agents will automate data cleaning, feature engineering, and model tuning, letting data scientists focus on strategy and problem-solving.
turboquant compresses llm caches to 3 bits
google's turboquant uses polarquant and qjl to cut kv cache memory by over 5x without retraining or accuracy loss.
adaptive importance sampling fixes quantized rl training
a new method corrects policy gradient bias from low-precision rollouts in llm reinforcement learning, preventing training collapse.
granite multilingual r2: compact 97m model tops sub-100m retrieval
ibm releases two apache 2.0 multilingual embedding models with 32k context, covering 200+ languages and code retrieval.
ai confidence alignment speeds human decision learning
study shows that matching ai confidence to human self-confidence helps people learn faster when making decisions with ai assistance.
asynchronous batching speeds up llm inference
separating cpu and gpu work with cuda streams and events reduces idle time, boosting throughput by up to 24%.
5 small language models for agentic tool calling
five compact open-weight language models that support structured tool calling for agentic ai workflows.
ocean concept bottleneck model explains heat predictions
a new model uses physics-based concepts to make ocean heat forecasts interpretable, revealing the drivers behind predictions.
learning when to act for safe, communication-efficient rl
a new method learns both control and when to communicate, using a safety shield to reduce sampling while maintaining stability.
quide: a single score for quantized model efficiency
a new metric called quide collapses compression, accuracy, and latency into one score to find optimal quantization levels for neural networks.
10 github repos to learn self-hosting
a curated list of github repositories that teach self-hosting skills from discovery to deployment, monitoring, and secure access.
pytorch 2.12: faster linalg, unified graphs, and mx export
pytorch 2.12 brings up to 100x faster batched eigendecomposition, a device-agnostic graph api, and support for exporting microscaling quantized models.
adaptive steering for discrete diffusion language models
uniform control schedules hurt text quality in discrete diffusion models, but a new adaptive scheduler improves multi-attribute steering by aligning interventions with each attribute's unique denoising timeline.
vertex-softmax tightens transformer verification
new method finds exact softmax bounds from score intervals, reducing slack in transformer robustness checks.
balora adds bayesian uncertainty to lora fine-tuning
balora extends lora with bayesian adaptation, improving accuracy and providing uncertainty estimates for large model fine-tuning.
python scripts for time series analysis
five reusable python scripts handle common time series tasks like resampling, anomaly detection, decomposition, forecasting, and multi-series comparison.
executorch labs for arm edge ai
arm releases hands-on jupyter labs to deploy pytorch models on edge cpus and npus using executorch.