Embedded Files
Organizers: David Pechersky, Artane Siad, Ning Su, Justin Yeh, Zhen Zhang, Hongyan Zhao
Time: Mondays 18:00 - 19:00 Location: Jingzhai
The goal of the Learning Machine Learning (LML!) seminar is to help participants build a conceptual foundation in machine learning by selected readings of the classical literature.
We will cover foundational work in reinforcement learning, large language models and world models, picking up current developments as they naturally arise along the way. The format will be one paper per week, read closely and discussed carefully. We will also engage with implementation — understanding how these systems are actually built.
We are interested in understanding machine learning clearly, conceptually, and from the ground up.
As a companion activity, join us for regular Sunday hackathons :
develop hands-on fluency by building.
Schedule
11 May 2026 – Justin Yeh, Tsinghua University
The Transformer Upgrade Path: 1. Tracing the origins of Sinusoidal Encoding
This talk goes into the mathematical details of Su Jianlin's blog on positional embedding to establish the foundations necessary for understanding and later building Rotary Positional Encoding (RoPE) in subsequent talks.
Resources: https://main-horse.github.io/translations/transformer-upgrade/, https://arxiv.org/pdf/2104.09864
20 April 2026 – Justin Yeh, Tsinghua University
From Seq2Seq to Transformer.
This talk introduces the foundational sequence-to-sequence (seq2seq) architecture and the attention mechanism that revolutionized it. We start with the encoder-decoder framework, covering training basics and simple models, then explain why attention is needed and how it works. From there, we build up to the Transformer, the modern workhorse of seq2seq. We also discuss practical essentials: subword segmentation (e.g., Byte Pair Encoding), inference methods like beam search, and finally touch on how we can analyze and interpret what these models have learned. The goal is to go through each component in technical detail, from the ground up making the ideas accessible to a beginner audience without glossing over how things actually work.
Resources: https://arxiv.org/pdf/1409.3215, https://lena-voita.github.io/nlp_course/seq2seq_and_attention.html
13 April 2026 – Marc Wegmann, Technische Universität München
Implementing Reinforcement Learning in Production Planning: a step-by-step methodology for industrial use cases.
As traditional production planning and control (PPC) struggles with increasing volatility and complexity, Reinforcement Learning (RL) offers a path toward more adaptive and robust decision-making. This presentation provides a goal-oriented guide on how to practically implement RL in industrial settings, using a structured methodology to bridge the gap between theory and application. We walk through the critical steps of the implementation process: from identifying high-potential use cases to designing the essential "building blocks" (action space, reward function, and state space) tailored to specific industrial settings. The session concludes by addressing practical hurdles such as the "sim-to-real" gap, changing environmental conditions, and the need for transparency in industrial AI systems.
12 April 2026 (11:00 am Beijing Time!) – Haocheng Ju and Guoxiong Gao, Peking University
AI-Powered Mathematical Research: from conjecture generation to proof.
Discussion session with PKU Ai4Math members about the resolution of Anderson's conjecture and the Rethlas/Archon architecture.
Resources: https://arxiv.org/pdf/2604.03789v1
30 March 2026 — Ning Su, YMSC and Tsinghua University
The transformer architecture and "Attention is All You Need".
This talk will introduce the transformer architecture and modern attention mechanisms underpinning modern LLMs following "Attention is all you need" by Vaswani et. al.
Resources: https://arxiv.org/pdf/1706.03762, https://bbycroft.net/llm
Page updated
Google Sites
Report abuse