Search this site
Embedded Files
Towards AGI
  • Home
  • Schedule
  • Topics&Papers
    • Adversarial Robustness
    • Alignment and Safety
    • CompPsych-FoMo
    • Compression and Fast Inference
    • Continual Learning at Scale
    • Emergence & Phase Transitions in ML
    • Foundation Models
    • Generalization (iid and ood)
    • High Performance Computing
    • Knowledge Fusion
    • Neural Scaling Laws
    • Out-of-Distribution Generalization
    • Scaling Laws in Nature
    • State Space Models
    • Time Series Foundation Models
Towards AGI
  • Home
  • Schedule
  • Topics&Papers
    • Adversarial Robustness
    • Alignment and Safety
    • CompPsych-FoMo
    • Compression and Fast Inference
    • Continual Learning at Scale
    • Emergence & Phase Transitions in ML
    • Foundation Models
    • Generalization (iid and ood)
    • High Performance Computing
    • Knowledge Fusion
    • Neural Scaling Laws
    • Out-of-Distribution Generalization
    • Scaling Laws in Nature
    • State Space Models
    • Time Series Foundation Models
  • More
    • Home
    • Schedule
    • Topics&Papers
      • Adversarial Robustness
      • Alignment and Safety
      • CompPsych-FoMo
      • Compression and Fast Inference
      • Continual Learning at Scale
      • Emergence & Phase Transitions in ML
      • Foundation Models
      • Generalization (iid and ood)
      • High Performance Computing
      • Knowledge Fusion
      • Neural Scaling Laws
      • Out-of-Distribution Generalization
      • Scaling Laws in Nature
      • State Space Models
      • Time Series Foundation Models

Schedule

Topics & Papers  

Paper presentations and projects: schedule & sign up sheet


Deadlines:
February - project proposals due

TBA - final project submission due

Class 1 (Wed, Jan 8, 4:30-6:30pm)

Lecturer: Irina Rish 

Topic: Intro and Overview: A brief history of AI at Scale  (slides)

Papers:  The Bitter Lesson,   GPT-3 paper: Language Models are Few-Shot Learners

Class 2  (Mon, Jan 13, 4:30-6:30pm)

Lecturer: Irina Rish 

Topic:  Intro and Overview: Continual Learning at Scale (slides, video)

Class 3  (Wed, Jan 15, 4:30-6:30pm)

Lecturer: Irina Rish 

Topic:  Overview of Emergence/Grokking, Large-Scale Projects, Time-Series FoMo (video)

 Time-Series Foundation Models
AI@Scale Workshop (including tutorial on HPC and distributed LLM training)

Class 4  (Mon, Jan 20, 4:30-6:30pm)

Lecturers: Irina Rish  and Tommaso Tossato

Topic:  Overview of Projects on Psych Eval of LLMs (video posted on discord)

Additional reading:  Computational Psychology & Foundation Models  papers  and   ComPsy FoMo Workshop

Class 5  (Wed, Jan 22, 4:30-6:30pm)

Part 1: Lecturer:  Daria Yasafova

Topic :  Scaling  Laws for Neural Language Models  (slides, video)


Additional video: Neural Scaling Laws and GPT-3

Class 6  (Mon, Jan 27, 4:30-6:30pm)

Lecturers: Irina Rish   

Topic:   Training Compute-Optimal Large Language Models  (Chinchilla Explained: video)

Additional reading: Scaling Laws for LLMs: from GPT-3 to o3

Class 7  (Wed, Jan 29, 4:30-6:30pm)

Part 1: Lecturer:  Prateek Humane

Topic:  DeepSeek-R1: Incentivizing reasoning capability in LLMs via reinforcement learning   (slides, video)


Additional Reading:
Open-R1: a fully open reproduction of DeepSeek-R1
DeepSeek-R1 explained (Medium article)

Class 8  (Mon, Feb 3, 4:30-6:30pm)

Part1:   Lecturers: Yorguin Jose Mantilla Ramos,  Shruti Bibra

Topic:  Scale Alone Does not Improve Mechanistic Interpretability in Vision Models  (slides, video)


Part 2: Lecturer: Daria Yasafova
Topic: Language models are few shot learners  (slides, video - part 2)


Additional reading: Scaling Laws for LLMs: from GPT-3 to o3

Class 9  (Wed, Feb 5, 4:30-6:30pm)

Part 1: Lecturer:  Prateek Humane

Topic:  Alignment faking in large language models  (slides, video)


Part 2: Lecturers:  Edward Habelrih  and Frederic Jarjour

Topic: Effect of scale on catastrophic forgetting in neural networks (slides, video - part 2)

Class 10  (Mon, Feb 10, 4:30-6:30pm)

Part1:   Lecturers: Michael Xi and Morteza Mahdiani

Topic:  Loss of plasticity in deep continual learning   (slides, video)


Part 2: Lecturer: Jama Hussein Mohamud
Topic:  Scaling Laws for Transfer   (slides, video - part 2)

Class 11  (Wed, Feb 12, 4:30-6:30pm)

Part 1: Lecturers:   Shruti Bibra and Yousef Kotp

Topic:  When Do We Not Need Larger Vision Models?  (slides, video)


Part 2: Lecturers:  Jiadi Yu and Mingze Li 

Topic: Zero-Shot Text-to-Image Generation (slides, video - part 2)

Class 12 (Mon, Feb 17, 4:30-6:30pm)

Part1:   Lecturers: Wenhao Xu and Huan Zhang

Topic:  PaLM: Scaling Language Modeling with Pathways   (slides, video)


Part 2: Lecturers:  Sungjae Cho  and   Anirudh Jamkhandi
Topic:  1. Scaling laws in the mammalian neocortex: does form provide clues to function?

2.  A Connectomic Hypothesis for the Hominization of the Brain   (slides, video - part 2)

Class 13  (Wed, Feb 19, 4:30-6:30pm)

Part 1: Lecturers:   Ivan Anokhin and Shahrad Mohammadzadeh

Topic:  RLHF: How to Learn from Human Feedback with Reinforcement Learning (slides, video)


Part 2: Lecturers:  Rishika Bhagwatkar and Aditya Sharma

Topic: RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning (slides, video - part 2)

Class 14 (Mon, Feb 24, 4:30-6:30pm)

Part1:   Lecturers: Sepehr Babapour and Alireza Dizaji

Topic:  Representation Projection Invariance Mitigates Representation Collapse   (slides, video)


Part 2: Lecturer:  Anas El Houssaini
Topic:  Open X-Embodiment: Robotic Learning Datasets and RT-XModels  (slides, video - part 2)

Class 15  (Wed, Feb 26, 4:30-6:30pm)

Part 1: Lecturers:   Azalée Robitaille and Artur Kuramshin

Topic:  Data scaling laws in imitation learning for robotic manipulation (slides, video)


Part 2: Lecturers:  Manping Li and Yan Zhang

Topic: Simple and Scalable Strategies to Continually Pre-train Large Language Models  (slides, video - part 2)

Class 16 (Mon, March  3, 4:30-6:30pm)

Part1:   Lecturers: Roger Creus

Topic:  Voyager: An Open-Ended Embodied Agent with Large Language Models   (slides, video)


Part 2: Lecturers: Wenhao Xu and  Huan Zhang
Topic:  Towards Understanding Sycophancy in Language Models (slides, video - part 2)

Class 17 (Wed, March 5, 4:30-6:30pm)

Part 1: Lecturers:   Ivan Anokhin and Navid Hassan Zadeh

Topic:  Investigating Continual Pretraining in Large Language Models: Insights and Implications (slides, video)


Part 2: Lecturers:  Yorguin Jose Mantilla Ramos and Yousef Kotp

Topic: Progress measures for grokking via mechanistic interpretability (slides, video - part 2)

Class 18 (Mon, March  10, 4:30-6:30pm)

Part1:   Lecturers: Kun Ni and Yuxing Tian

Topic:  DeepSeek LLM: Scaling Open-Source Language Models with Longtermism  (slides, video)


Part 2: Lecturers: Jama Hussein Mohamud
Topic:  Emergent Abilities of Large Language Models  (slides, video - part 2)

Class 19 (Wed, March 12, 4:30-6:30pm)

Part 1: Lecturers:   Jia Ao Sun and Hao Yu

Topic:  Constitutional AI: Harmlessness from AI Feedback (slides, video)


Part 2: Lecturers:  Sungjae Cho  and  Anirudh Jamkhandi

Topic: Brain-inspired replay for continual learning with artificial neural networks (slides, video - part 2)

Class 20 (Mon, March  17, 4:30-6:30pm)

Part1:   Lecturers: Rishika Bhagwatkar and Aditya Sharma

Topic:  Byte Latent Transformer: Patches Scale Better Than Tokens  (slides, video)

Class 21  (Wed, March 19, 4:30-6:30pm)

Part 1: Lecturers:   Azalée Robitaille and Artur Kuramshin

Topic:  Robotic Control via Embodied Chain-of-Thought Reasoning (slides, video)


Part 2: Lecturers:  Edward Habelrih  and  Frederic Jarjour

Topic:   Large Language Models can Strategically Deceive their Users when Put Under Pressure (slides, video - part 2)

Class 22  (Mon, March  24, 4:30-6:30pm)

Part1:   Lecturers: Kun Ni and Zibo Shang

Topic:  Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters  (slides, video)


Part 2: Lecturers: Anirudh Buvanesh and  Ankur Sikarwar
Topic:  The Platonic Representation Hypothesis (slides, video - part 2)

Class 23 (Wed, March 26, 4:30-6:30pm)

Part 1: Lecturers:   Juan David Guerra and Mauricio Rivera

Topic:  Rho-1: Not All Tokens Are What You Need (slides, video)


Part 2: Lecturers:  Sepehr Babapour

Topic:   Leveraging clinical data across healthcare institutions for continual learning of predictive risk models (slides, video - part 2)

Class 24  (Mon, March  31, 4:30-6:30pm)

Part1:   Lecturers:  Anirudh Buvanesh and Ayush Agrawal

Topic:  Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs (slides, video)


Part 2: Lecturers: Alexandrine Fortier and  Mikhail Kamara
Topic:  BrainWash: A Poisoning Attack to Forget in Continual Learning (slides, video - part 2)

Class 25 (Wed, April 2, 4:30-6:30pm)

Part 1: Lecturers:    Jiadi Yu and Mingze Li

Topic:  s1: Simple test-time scaling (slides, video)


Part 2: Lecturers:  Manping Li and Yan Zhang

Topic:   Learning Transferable Visual Models From Natural Language Supervision (slides, video - part 2)

Class 26  (Mon, April 7, 4:30-6:30pm)

Part1:   Lecturers:  Samin Mahdipour and Mariem ben Slimen

Topic:  Toward Next-Generation Artificial Intelligence: Catalyzing the NeuroAI Revolution (slides, video)


Class 27 (Wed, April 9, 4:30-6:30pm)

Part 1: Lecturers:    Ankur Sikarwar and Ayush Agrawal

Topic:  Genie: Generative Interactive Environments (slides, video)


Part 2: Lecturers:  Michael Xi and Morteza Mahdiani

Topic:   Scaling laws for decoding images from brain activity (slides, video - part 2)

Class 28  (Mon, April  14, 4:30-6:30pm)

Part 1: Lecturers: Alexandrine Fortier and  Mikhail Kamara
Topic:  Deception abilities emerged in large language models  (slides, video - part 2)

Class 29 (Wed, April 16, 4:30-6:30pm)

Part 1: Lecturers:    Mauricio Rivera and Juan David Guerra

Topic:  Dynamic Neural Regeneration (slides, video)


Part 2: Lecturers:  Navid Hassan Zadeh and Shahrad Mohammadzadeh

Topic:   Chronos: Learning the Language of Time Series  (slides, video - part 2)

Class 30 (Mon, April  21, 4:30-6:30pm)

Part 1: Lecturers: Yuxing Tian and  Zibo Shang
Topic:  Reasoning Models Don’t Always Say What They Think  (slides, video - part 2)


Part 2: Lecturers:  Artiom Matvei and  William Chidiac

Topic:   Mixtures of Experts Unlock Parameter Scaling for Deep RL  (slides, video - part 2)

Class 31 (Wed, April 23, 4:30-6:30pm)

Part 1: Lecturers:    Yajie Luo and Yihong Wu

Topic:  SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training  (slides, video)


Part 2: Lecturers:  Roger Creus

Topic:   Mastering Board Games by External and Internal Planning with Language Models  (slides, video - part 2)

Course Project Presentations: Poster Session on Monday, May 5th, 1pm - 4:00pm

Google Sites
Report abuse
Page details
Page updated
Google Sites
Report abuse