1. 1. Introduction
  3. Research Methodologies
  4. 2. Course
  5. 3. Seminar 1
  6. 4. Seminar 2
  7. 5. Extras
  9. Models
  10. 6. GNNs
  11. 7. ViTs
  12. 8. General Circulation Models
  13. 9. Structured State Spaces
  15. Transformer Variants
  16. 10. MaxViT
  17. 11. HieraViT
  18. 12. Swin Transformer
  19. 13. Swin Transformer V2
  20. 14. Spatio-Temporal Swin-Transformer
  21. 15. Multiscaled stacked ViT
  22. 16. Efficient Transformers
    1. 16.1. Linear Transformers
    2. 16.2. Linear Transformers as FWP
    3. 16.3. The Devil in Linear Transformers
    4. 16.4. Performer
    5. 16.5. SRFormer
  24. 17. Test time training
  25. 18. TTT++
  26. 19. Universal Transformers
  28. Weather Models
  29. 20. Rise of Data-Driven Weather Forecasting
  30. 21. ClimaX
  31. 22. PanguWeather
  32. 23. MetNet
  33. 24. FuXi
  34. 25. FourcastNet
  35. 26. GraphCast
  36. 27. PrithViWxC
  37. 28. ORBIT
  38. 29. Climate Benchmarks
    1. 29.1. ClimateLearn
    2. 29.2. ClimateBench
    3. 29.3. WeatherBench
  40. 30. IPCC Reports
    1. 30.1. WG1
    2. 30.2. WG2
    3. 30.3. WG3
    4. 30.4. AR6
  43. Misc and Domain Specific Models
  44. 31. ACE
  45. 32. ExPT
  46. 33. Closed-form continuous-time neural networks
  47. 34. Recurrent Fast Weight Programmers
  48. 35. Liquid Neural Networks
  49. 36. Liquid S4
  50. 37. AI CCA
  51. 38. FireCast
  52. 39. FarSite
  53. 40. Wildfire risk modeling
  54. 41. XAI wildfire in the Mediterranean countries
  56. Scaling Benchmarks
  57. 42. Training LLMs on leadership‐class supercomputers
  58. 43. Scaling Laws for LLMS
  59. 44. Compute Optimal LLMs
  60. 45. Data Optimal LLMs
  62. Development
  63. 46. Parallelization
  65. Boring AI
  66. 47. Data Quality
  67. 48. Trustworthy AI
  68. 49. Provenance Documentation for AI
  69. 50. Information Required for XAI
  70. 51. Explainable AI Techniques
  71. 52. Provenance
    1. 52.1. Open Provenance Model
    2. 52.2. PROV Notation
    3. 52.3. PROV Constraints
    4. 52.4. PROV-JSON
    5. 52.5. A Graph Model for Data Provenance
  73. 53. Provenance Supporting Hyperparameter Analysis in DNNs
  74. 54. FAIR Guiding Principles
  75. 55. Advances, Challenges, and Opportunities in TAI
  76. 56. Provenance in ML
    1. 56.1. Provenance in ML Lifecycle
    2. 56.2. Prov supporting Hyperparameter Analysis in DNNs
    3. 56.3. Data Provenance for Distributed ML
    4. 56.4. Linking PROV and XAI in distributed ML
    5. 56.5. MLFlow2PROV
    6. 56.6. LIMA
    7. 56.7. ProvLake
    8. 56.8. Efficient runtime capture of multiworkflow data using prov
    9. 56.9. PrIU
    10. 56.10. Provenance for Responsible AI Systems
    11. 56.11. Provenance Documentation for XAI
      1. 56.11.1. Data Provenance Initiative
  78. 57. RO-Crate
  79. 58. Workflow Run ROC
  80. 59. Collecting Managing and Analyzing Provenance Data
  81. 60. Provenance-based IDs
  83. Energy Efficiency
  84. 61. Metrics
  85. 62. CarbonLLM
  86. 63. HVAC
  87. 64. Computational/Energy Use of CMIP6
  88. 65. Energy Intensity of Internet Data Transmission
  89. 66. Green AI
  90. 67. Green 500 Metrics
  91. 68. Data center EE
  93. Growing Models
  94. 69. LSTMVis
  95. 70. H-State and Semantic Structure
  96. 71. Hidden Memories of RNNs
  97. 72. Analysis of Entropy in Hidden States
  99. Useful Stuff
  100. 73. Notes for Talks
    1. 73.1. A Data-Oriented Perspective
    2. 73.2. Ai for Good talk
  102. 74. Tropical Cyclones
  103. 75. Model Comparison
  104. 76. Challenges and Opportunities
  105. 77. FM4C Improvements
  106. 78. WandB Webinar
  107. 79. Zero
  108. 80. Zero++
  110. Ideas
  111. 81. Fire behavior
  112. 82. ML for Climate
  113. 83. Incremental Models
  114. 84. Scaling Issues
  116. Footnotes

Foundation Model 4 Climate Notes