| Lecture 1 |
-- |
Topics: (no slides)
- Formal introduction
- Course details
- Syllabus
|
|
|
| Lecture 2 |
-- |
Topics: (slides)
- Recommender landscape personalization, retrieval vs ranking, cold start, feedback loops.
|
|
|
| Lecture 3 |
-- |
Topics: (slides)
- Data & problem setup implicit vs explicit feedback; sessions; negatives; train/val/test splits.
|
|
|
| Lecture 4 |
-- |
Topics: (slides)
- Baselines popularity, recency, simple heuristics; why baselines are hard to beat
|
|
|
| Lecture 5 |
-- |
Topics: (slides)
- Baselines popularity, recency, simple heuristics; why baselines are hard to beat
|
|
|
| Lecture 6 |
-- |
Topics: (slides)
- Evaluation Precision@K/Recall@K/NDCG/MAP; offline vs online; leakage pitfalls
|
|
|
| Lecture 7 |
-- |
Topics: (slides)
- Evaluation Precision@K/Recall@K/NDCG/MAP; offline vs online; leakage pitfalls
|
|
|
| Lecture 8 |
-- |
Topics: (slides)
- Classical CF user-user/item-item similarity; cosine/pearson; scalability issues.
|
|
|
| Lecture 9 |
-- |
Topics: (slides)
- Classical CF user-user/item-item similarity; cosine/pearson; scalability issues.
|
|
|
| Lecture 10 |
-- |
Topics: (slides)
- Matrix factorization (MF) latent factors; SGD/ALS; regularization; implicit MF (BPR intuition)
|
|
|
| Lecture 11 |
-- |
Topics: (slides)
- Matrix factorization (MF) latent factors; SGD/ALS; regularization; implicit MF (BPR intuition)
|
|
|
| Lecture 12 |
-- |
Topics: (slides)
- Matrix factorization (MF) latent factors; SGD/ALS; regularization; implicit MF (BPR intuition)
|
|
|
| Lecture 13 |
-- |
Topics: (slides)
- Matrix factorization (MF) latent factors; SGD/ALS; regularization; implicit MF (BPR intuition)
|
|
|
| Lecture 14 |
-- |
Topics: (slides)
- Learning-to-rank basics pointwise/pairwise/listwise; negative sampling strategies
|
|
|
| Lecture 15 |
-- |
Topics: (slides)
- Learning-to-rank basics pointwise/pairwise/listwise; negative sampling strategies
|
|
|
| Lecture 16 |
-- |
Topics: (slides)
- Neural CF MLP interactions, embeddings, feature crosses; implementation
|
|
|
| Lecture 17 |
-- |
Topics: (slides)
- Neural CF MLP interactions, embeddings, feature crosses; implementation
|
|
|
| Lecture 18 |
-- |
Topics: (slides)
- Neural CF MLP interactions, embeddings, feature crosses; implementation
|
|
|
| Lecture 19 |
-- |
Topics: (slides)
- Neural CF MLP interactions, embeddings, feature crosses; implementation
|
|
|
| Lecture 20 |
-- |
Topics: (slides)
- Wide & Deep / DeepFM intuition combining memorization + generalization; structured features
|
|
|
|
-- |
-- |
(Feb 25) Last Date for Proposal Submission. |
|
|
-- |
-- |
Mid Semester Exam Week |
Best of Luck.
|
| Lecture 21 |
-- |
Topics: (slides)
- Wide & Deep / DeepFM intuition combining memorization + generalization; structured features
|
|
|
| Lecture 22 |
-- |
Topics: (slides)
- Two-tower retrieval models dual encoders; approximate nearest neighbor; candidate generation
|
|
|
| Lecture 23 |
-- |
Topics: (slides)
- Two-tower retrieval models dual encoders; approximate nearest neighbor; candidate generation
|
|
|
| Lecture 24 |
-- |
Topics: (slides)
- Ranking models DNN rankers, context features; calibration; diversity constraints
|
|
|
| Lecture 25 |
-- |
Topics: (slides)
- Ranking models DNN rankers, context features; calibration; diversity constraints
|
|
|
| Lecture 26 |
-- |
Topics: (slides)
- Sequence recommenders I Markov chains → RNNs; session-based recs
|
|
|
| Lecture 27 |
-- |
Topics: (slides)
- Sequence recommenders I Markov chains → RNNs; session-based recs
|
|
|
| Lecture 28 |
-- |
Topics: (slides)
- Sequence recommenders II Transformers for rec (SASRec-style ideas); positional encoding for sessions
|
|
|
| Lecture 29 |
-- |
Topics: (slides)
- Sequence recommenders II Transformers for rec (SASRec-style ideas); positional encoding for sessions
|
|
|
| Lecture 30 |
-- |
Topics: (slides)
- Graph recommenders (intro) bipartite graphs, message passing intuition; when it helps
|
|
|
| Lecture 31 |
-- |
Topics: (slides)
- Graph recommenders (intro) bipartite graphs, message passing intuition; when it helps
|
|
|
| Lecture 32 |
-- |
Topics: (slides)
- Cold start strategies content-based, hybrid, metadata embeddings, zero-shot user modeling
|
|
|
|
-- |
-- |
(Mar 30) - Last Date of Mid Presentation. |
|
| Lecture 33 |
-- |
Topics: (slides)
- Cold start strategies content-based, hybrid, metadata embeddings, zero-shot user modeling
|
|
|
| Lecture 34 |
-- |
Topics: (slides)
- Bias & fairness popularity bias, exposure bias, filter bubbles; debiasing ideas
|
|
|
| Lecture 35 |
-- |
Topics: (slides)
- Bias & fairness popularity bias, exposure bias, filter bubbles; debiasing ideas
|
|
|
| Lecture 36 |
-- |
Topics: (slides)
- Exploration & bandits epsilon-greedy, UCB, Thompson sampling; counterfactual logging basics
|
|
|
| Lecture 37 |
-- |
Topics: (slides)
- Serving & systems feature stores, ANN indexes, latency budgets; batch vs real-time features
|
|
|
| Lecture 38 |
-- |
Topics: (slides)
- Monitoring drift, feedback loops, A/B testing guardrails; metric dashboards
|
|
|
| Lecture 39 |
-- |
Topics: (slides)
- Capstone presentations full pipeline (retrieval + ranker) + evaluation report
|
|
|
| Lecture 40 |
-- |
Topics: (slides)
- Capstone presentations full pipeline (retrieval + ranker) + evaluation report
|
|
|
| Lecture 41 |
-- |
Topics: (slides)
|
|
|
| Lecture 42 |
-- |
Topics: (slides)
|
|
|
|
-- |
-- |
(Apl 30) Last Date of Codes Submission. |
|
|
-- |
-- |
(Apl 30) Dead Line for Final Presentation Video Submission. |
|
|
-- |
-- |
(Apl 30) Dead Line for Report Submission. |
|
|
-- |
-- |
End Semester Exam Week |
Best of Luck.
|
|
Link Added on Last Date for Submission :
|
DA626
