🤖 AI - ibeyst

Hierarchical Reinforcement Learning for Sparse-Reward Search in Commutative Algebra

Escaping the chain-of-thought trap: What is next for LLM reasoning

Covers 2 stories including Uber burned through its entire 2026 AI budget in four months. Now its COO is questioning whether it’s worth it

Discussed on Substack

⚙️Mechanical Sympathy Cryptology ePrint Archive·

BootNet: Homomorphic CNN Inference with Convolution and ReLU Fused in Bootstrapping

🌐Distributed Systems medium.com

InceptionNet Explained: The Architecture That Taught CNNs to Think at Multiple Scales

⚙️Mechanical Sympathy medium.com

Will AI Eventually Become Cheap? The Surprising Economics of Falling Compute Costs

🌳LSM Trees arXiv·

Leveraging AutoML for Sustainable Deep Learning: A Multi-Objective HPO Approach on Deep Shift Neural Networks

⚙️Mechanical Sympathy Nature

Optical metasurfaces for general vision processing on the edge

🔄Differential Dataflow arXiv·

Comparative Evaluation of Machine Learning and Deep Learning Models for Wound-Rotor Synchronous Motor Performance Prediction

🧠Query Planners arXiv·

Early-Exit Graph Neural Networks for Link Prediction

🗄️Databases medium.com

The Most Common Activation Functions Used in Neural Networks

🌳LSM Trees arXiv·

Field-level weak lensing cosmology with $<100$ simulations using multifidelity simulation-based inference

⌨️Programming arXiv·

Breaking chains with trees: Deep learning with $\mathcal{O}(\log N)$ parallel time complexity

🔄Differential Dataflow medium.com

Forward Propagation: How Neural Networks Process Data

🧠Query Planners arXiv·

The New Associationism: Lessons from Deep Learning

⚙️Mechanical Sympathy arXiv·

Efficient Network Inference via Hardware-Aware Architecture Search, Model Pruning & Quantization

🌐Distributed Systems arXiv·

Deep Learning (Part-02): Basics of Deep Learning & Neural Networks

Inside ResNet50: A Layer-by-Layer Deep Dive

From Scratch: Implementing the Transformer Architecture in Python

Munich 1991: The Roots of the Current AI Boom

Hierarchical Reinforcement Learning for Sparse-Reward Search in Commutative Algebra

Escaping the chain-of-thought trap: What is next for LLM reasoning

BootNet: Homomorphic CNN Inference with Convolution and ReLU Fused in Bootstrapping

InceptionNet Explained: The Architecture That Taught CNNs to Think at Multiple Scales

Will AI Eventually Become Cheap? The Surprising Economics of Falling Compute Costs

Leveraging AutoML for Sustainable Deep Learning: A Multi-Objective HPO Approach on Deep Shift Neural Networks

Optical metasurfaces for general vision processing on the edge

Comparative Evaluation of Machine Learning and Deep Learning Models for Wound-Rotor Synchronous Motor Performance Prediction

Early-Exit Graph Neural Networks for Link Prediction

The Most Common Activation Functions Used in Neural Networks

Field-level weak lensing cosmology with $<100$ simulations using multifidelity simulation-based inference

Breaking chains with trees: Deep learning with $\mathcal{O}(\log N)$ parallel time complexity

Forward Propagation: How Neural Networks Process Data

The New Associationism: Lessons from Deep Learning

Efficient Network Inference via Hardware-Aware Architecture Search, Model Pruning & Quantization

Scalable Physics-Inspired Transformers for Spin Glasses