Schedule

The schedule may be adjusted during the course. All deadlines are at 11:59pm (CT). Paper reviews are due the night before class, and paper presentation slides are due three days before the presentation. See the syllabus for more details.

Week	Date	Topic	Readings & Action Items
#1	Aug 26 Wed	Class 1 Course overview	Optional How to Read a Paper (SIGCOMM CCR ’07) Deadline Submit topic preferences for paper presentation
	Aug 28 Fri	Class 2 ML-for-systems taxonomy	Required A Taxonomy of ML for Systems Problems (IEEE Micro ’20) Optional A View on Deep Reinforcement Learning in System Optimization (arXiv ’19)
#2	Sep 2 Wed	Class 3 A refresher on core ML models and algorithms
	Sep 4 Fri	Class 4 The classic: learned indexes	Required The Case for Learned Index Structures (SIGMOD ’18) Optional Towards instance-optimized data systems (VLDB ’21)
	Sep 6 Sun		Deadline Submit project team roster
#3	Sep 9 Wed	Class 5 The latest: AI agents for research	Required AlphaEvolve: A coding agent for scientific and algorithmic discovery (arXiv ’25) Optional Barbarians at the Gate: How AI is Upending Systems Research (arXiv ’25) Optional MetaMuse: Algorithm Generation via Creative Ideation (ICLR ’26) Optional Glia: A Human-Inspired AI for Automated Systems Design and Optimization (CAIS ’26)
	Sep 11 Fri	Class 6 Attendance Required AI agents for cloud operations: Guest lecture by Yinfang Chen	Optional STRATUS: A Multi-agent System for Autonomous Reliability Engineering of Modern Clouds (NeurIPS ’25) Optional Automatic Root Cause Analysis via Large Language Models for Cloud Incidents (EuroSys ’24)
	Sep 13 Sun		Deadline Submit project proposal
#4	Sep 16 Wed	Class 7 Attendance Required AI agents for network management: Guest lecture by Kevin Hsieh	Optional MeshAgent: Enabling Reliable Network Management with Large Language Models (SIGMETRICS ’26) Optional NetArena: Dynamic Benchmarks for AI Agents in Network Automation (ICLR ’26) Optional Intent-Driven Network Management with Multi-Agent LLMs: The Confucius Framework (SIGCOMM ’25)
	Sep 18 Fri	Class 8 AI agents for video analytics	Required Ava: Towards Agentic Video Analytics with Vision Language Models (NSDI ’26) Optional OmAgent: A Multi-modal Agent Framework for Complex Video Understanding with Task Divide-and-Conquer (EMNLP ’24) Optional VCA: Video Curious Agent for Long Video Understanding (ICCV ’25)
#5	Sep 23 Wed	Class 9 ML-based anomaly detection	Required Gandalf: An Intelligent, End-To-End Analytics Service for Safe Deployment in Large-Scale Cloud Infrastructure (NSDI ’20) Optional Predictive and Adaptive Failure Mitigation to Avert Production Cloud VM Interruptions (OSDI ’20)
	Sep 25 Fri	Class 10 Configuration autotuning	Required CherryPick: Adaptively Unearthing the Best Cloud Configurations for Big Data Analytics (NSDI ’17) Optional Configanator: A Data-driven Approach to Improving CDN Performance (NSDI ’22)
#6	Sep 30 Wed	Class 11 Cloud resource management with ML	Required Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms (SOSP ’17) Optional Virtual Machine Allocation with Lifetime Predictions (MLSys ’23) Optional LAVA: Lifetime-Aware VM Allocation with Learned Distributions and Adaptation to Mispredictions (MLSys ’25)
	Oct 2 Fri	Class 12 Microservice resource management with ML	Required FIRM: An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices (OSDI ’20) Optional Sinan: ML-Based and QoS-Aware Resource Management for Cloud Microservices (ASPLOS ’21) Optional Autothrottle: A Practical Bi-Level Approach to Resource Management for SLO-Targeted Microservices (NSDI ’24)
	Oct 4 Sun		Deadline Submit the first individual progress update
#7	Oct 7 Wed	Class 13 Online learned congestion control	Required PCC Vivace: Online-Learning Congestion Control (NSDI ’18) Optional OnRL: Improving Mobile Video Telephony via Online Reinforcement Learning (MobiCom ’20)
	Oct 9 Fri	Class 14 Offline learned congestion control	Required Computers Can Learn from the Heuristic Designs and Master Internet Congestion Control (SIGCOMM ’23) Optional Mowgli: Passively Learned Rate Control for Real-Time Video (NSDI ’25)
#8	Oct 14 Wed	Class 15 ML-based video streaming	Required Neural Adaptive Video Streaming with Pensieve (SIGCOMM ’17) Optional Learning in situ: a randomized experiment in video streaming (NSDI ’20)
	Oct 16 Fri	Class 16 ML-based traffic engineering	Required DOTE: Rethinking (Predictive) WAN Traffic Engineering (NSDI ’23) Optional Teal: Learning-Accelerated Optimization of WAN Traffic Engineering (SIGCOMM ’23)
#9	Oct 21 Wed	Class 17 Learned cache replacement	Required Learning Relaxed Belady for Content Distribution Network Caching (NSDI ’20) Optional Raven: Belady-Guided, Predictive (Deep) Learning for In-Memory and Content Caching (CoNEXT ’22) Optional 3L-Cache: Low Overhead and Precise Learning-based Eviction Policy for Caches (FAST ’25)
	Oct 23 Fri	Class 18 Combining heuristics and ML for cache replacement	Required HALP: Heuristic Aided Learned Preference Eviction Policy for YouTube Content Delivery Network (NSDI ’23) Optional Learning Cache Replacement with Cacheus (FAST ’21) Optional Learning-Augmented Heuristics: Simple yet Smart, Robust and Interpretable Cache Eviction (OSDI ’26)
	Oct 25 Sun		Deadline Submit the second individual progress update
#10	Oct 28 Wed	Class 19 Generative AI for real-time video	Required Gemino: Practical and Robust Neural Compression for Video Conferencing (NSDI ’24) Optional Grace: Loss-Resilient Real-Time Video through Neural Codecs (NSDI ’24)
	Oct 30 Fri	Class 20 Generative AI for trace synthesis	Required Practical GAN-based Synthetic IP Header Trace Generation using NetShare (SIGCOMM ’22) Optional Making Logic a First-Class Citizen in Generative ML for Networking (NSDI ’26)
#11	Nov 4 Wed	Class 21 ML inside OS	Required LinnOS: Predictability on Unpredictable Flash Storage with a Light Neural Network (OSDI ’20) Optional Learning-based Memory Allocation for C++ Server Workloads (ASPLOS ’20)
	Nov 6 Fri	Class 22 Causal inference and ML	Required Sage: Practical and Scalable ML-Driven Performance Debugging in Microservices (ASPLOS ’21) Optional CausalSim: A Causal Framework for Unbiased Trace-Driven Simulation (NSDI ’23)
#12	Nov 11 Wed	Class 23 Safety	Required Verifying Learning-Augmented Systems (SIGCOMM ’21) Optional Building Verified Neural Networks for Computer Systems with Ouroboros (MLSys ’23)
	Nov 13 Fri	Class 24 Robustness	Required AdvNet: Revealing Performance Issues in Network Protocols by Generating Adversarial Environments (CoNEXT ’26) Optional Genet: Automatic Curriculum Generation for Learning Adaptation in Networking (SIGCOMM ’22)
	Nov 15 Sun		Deadline Submit the last individual progress update
#13	Nov 18 Wed	Class 25 Generality	Required On the Use of ML for Blackbox System Performance Prediction (NSDI ’21) Optional FLASH: Fast Model Adaptation in ML-Centric Cloud Platforms (MLSys ’24)
	Nov 20 Fri	Class 26 Explainability	Required Agua: A Concept-Based Explainer for Learning-Enabled Systems (SIGCOMM ’25) Optional Interpreting Deep Learning-Based Networking Systems (SIGCOMM ’20)
#14	Nov 25 Wed	No class — Fall break
	Nov 27 Fri	No class — Fall break
#15	Dec 2 Wed	Class 27 Attendance Required Final presentation
	Dec 4 Fri	Class 28 Attendance Required Final presentation
#16	Dec 9 Wed	Class 29 Attendance Required AI agents for code optimization: Guest lecture by Martin Maas	Optional ECO: An AI-Driven Code Efficiency Optimizer for Warehouse Scale Computers (OSDI ’26)
#17	Dec 17 Thu		Deadline Submit final paper