Source: https://zachwills.net/how-to-use-claude-code-subagents-to-parallelize-development/

This book offers a comprehensive and accessible introduction to game theory, emphasizing both noncooperative and cooperative aspects of strategic decision-making. In the chapters on noncooperative game theory, you will explore advanced topics such as perfect equilibrium, evolutionary stable strategies, and correlated equilibrium, along with a range of subjects often underrepresented in other textbooks.The cooperative game theory sections cover essential topics like coalitional games, cake-cutting and fairness, cooperative bargaining, and matching theory. Additionally, the book includes an insightful chapter on mechanism design.Designed for use in one-semester advanced undergraduate or graduate-level courses, this textbook stands apart from others at the same level. Each chapter begins with clear theoretical definitions, followed by carefully detailed examples. Select chapters include propositions that either demonstrate the existence of equilibrium in abstract games or interrelate various game-theoretic concepts.While rigorous in its scope, the book assumes no advanced background in calculus or algebra. The mathematical exposition is kept as straightforward and self-contained as possible, ensuring that readers can easily apply theoretical ideas to practical examples and follow proofs with ease.

Problem: #

create pair of of people with similar properties. Minimize difference between members of a pair

Solution #

Algorithm for optimal node pairing that minimizes the total distance between paired nodes. This is essentially a minimum-weight perfect matching problem.I’ve created a comprehensive implementation of optimal node pairing algorithms. Here’s what the code includes:

Key Components: #

1. Node Class: Represents a point in 2D space with distance calculation
2. Multiple Algorithm Approaches:
   • Greedy: Fast O(n³) algorithm that repeatedly finds closest pairs
   • Nearest Neighbor: O(n²) algorithm that pairs each node with its nearest unpaired neighbor
   • Approximation with Local Search: Improves greedy solution using 2-opt style swaps
   • Brute Force Optimal: Guaranteed optimal solution via backtracking (exponential time)

Algorithm Trade-offs: #

• Greedy: Fast but may get stuck in local optima
• Nearest Neighbor: Faster, good for large datasets, decent quality
• Approximation: Best balance of speed and quality for medium datasets
• Brute Force: Perfect solution but only practical for small instances (≤8-10 nodes)

Usage Recommendations: #

• Small datasets (≤8 nodes): Use brute force for guaranteed optimal
• Medium datasets (10-100 nodes): Use approximation algorithm
• Large datasets (>100 nodes): Use nearest neighbor or greedy

The code includes a test function that demonstrates all algorithms on a sample dataset, showing the total distance and individual pairs for each approach. You can easily adapt it for your specific coordinate system or distance metric.

Mengukur detak jantung menggunakan WiFi. Pakai ESP32 dan Raspberry Pi juga bisa. https://news.ucsc.edu/2025/09/pulse-fi-wifi-heart-rate/ papernya:

• https://news.ucsc.edu/wp-content/uploads/2025/08/Pulse_Fi_short_paper_iotdcoss.pdf
• https://bpb-us-e1.wpmucdn.com/sites.ucsc.edu/dist/d/1595/files/2025/05/Pulse_Fi_short_paper_iotdcoss.pdf

Buku ini berisi referensi untuk kuliah pemrograman bahasa assembly untuk tingkat perguruan tinggi. Set instruksi yang dibahasa adalah x86-64 untuk prosesor x86-64 menggunakan sistem operasi Ubuntu 64 bit.

Kode di buku ini diuji di Ubuntu versi 22.04 LTS

Source:

• http://www.egr.unlv.edu/~ed/assembly64.pdf
• mirror

Today we’re releasing new @AnthropicAI research on how educators use AI, analyzing ~74,000 conversations from professors using @claudeai in collaboration with Northeastern University.

4 initial findings…

1. Educators are builders, not just users of AI. Faculty are creating interactive chemistry simulations, grading rubrics, and data dashboards with Claude Artifacts.
2. Educators automate the drudgery while staying in the loop for almost everything else. 77% of teaching and classroom instruction uses are collaborative, while 65% of financial/fundraising tasks are fully delegated. High-touch educational work remains human-centered.
3. Notable tension in the data: 49% of grading conversations showed automation patterns, yet faculty rated this as AI’s least effective application. This disconnect highlights ongoing debates around appropriate AI use in assessment.
4. AI is forcing pedagogical change. Professors are completely redesigning assessments—one shared: “I will never again assign a traditional research paper,” instead creating assignments requiring critical thinking even with AI assistance.

Abstrak

Abstract: We present Jet-Nemotron, a new family of hybrid-architecture language models, which matches or exceeds the accuracy of leading full-attention models while significantly improving generation throughput. Jet-Nemotron is developed using Post Neural Architecture Search (PostNAS), a novel neural architecture exploration pipeline that enables efficient model design. Unlike prior approaches, PostNAS begins with a pre-trained full-attention model and freezes its MLP weights, allowing efficient exploration of attention block designs. The pipeline includes four key components: (1) learning optimal full-attention layer placement and elimination, (2) linear attention block selection, (3) designing new attention blocks, and (4) performing hardware-aware hyperparameter search. Our Jet-Nemotron-2B model achieves comparable or superior accuracy to Qwen3, Qwen2.5, Gemma3, and Llama3.2 across a comprehensive suite of benchmarks while delivering up to 53.6× generation throughput speedup and 6.1× prefilling speedup. It also achieves higher accuracy on MMLU and MMLU-Pro than recent advanced MoE full-attention models, such as DeepSeek-V3-Small and Moonlight, despite their larger scale with 15B total and 2.2B activated parameters.

Microsoft just released VibeVoice - 1.5B SoTA Text to Speech model - MIT Licensed

It can generate up 90 minutes of audio Supports simultaneous generation of > 4 speakers Streaming and larger 7B model in-coming Capable of cross-lingual and singing synthesis

Source:

• https://huggingface.co/microsoft/VibeVoice-1.5B

Fakta-fakta: #

• Fact 1: Employment for young workers has declined in AI-exposed occupations
• Fact 2: Though overall employment continues to grow, employment growth for young workers in particular has been stagnant
• Fact 3: Entry-level employment has declined in applications of AI that automate work, with muted changes for augmentation
• Fact 4: Employment declines for young, AI-exposed workers remain after conditioning on firm-time shocks
• Fact 5: Labor market adjustments are visible in employment more than compensation
• Fact 6: Findings are largely consistent under alternative sample constructions

Source: https://github.com/murtaza-nasir/maestro