Implementation of a Hybrid Two-Tower Architecture for Music Recommendation. Solves cold-start problems by integrating content-based multimodal encoders with sequential user modeling.
Recipe recommender using Two-Tower retrieval, LightGBM ranking, and MMR diversification
Production-grade ML recommendation engine: Kafka → Flink → Two-Tower PyTorch + FAISS ANN + LinUCB Bandit → sub-100ms FastAPI serving with live React dashboard
Production-ready personalized recommender: two-tower retrieval, CatBoost ranking, optional LLM rerank. FastAPI + Qdrant + MLflow. Workshop by learnwithparam.com
This repo is made for Football analytics and is based on FIFA 24 data
Two-tower recommender (users - crypto tokens) in PyTorch, with a feature-based user tower and cold-start.
Production-grade MovieLens-25M recommender: two-stage retrieval + re-ranking with a full MLOps lifecycle (Spark features, MLflow, FAISS, FastAPI on Kubernetes, A/B testing, drift detection, Prometheus/Grafana).
A distributed ML recommendation system — real-time streaming, multi-node distributed training, and fault-tolerant, scalable serving.
Two-tower dual-encoder recommendation/retrieval with in-batch negatives and FAISS ANN (Recall@10/50, MRR)
🤖 Explore and optimize rewards with Bandexa, a PyTorch-native library for Neural-Linear Thompson Sampling in contextual bandits.
Short-video recommendation reproduction: Two-Tower recall, Faiss TopK, time split, negative sampling, Recall@50/NDCG@50, ablation and badcases.
2-Stage Recommendation on MovieLens-1M.
Production-grade multi-stage recommendation and personalization platform: candidate retrieval, two-tower neural embeddings, learning-to-rank, experimentation, serving, caching and monitoring.
A recommender on 25M real MovieLens ratings: temporal split, two-tower retrieval, and the logQ correction that took it from losing to a popularity counter (0.018) to beating every baseline (0.085), served through this series' own HNSW
Production-oriented personalized news retrieval and ranking on Microsoft MIND
Movie recommender: TensorFlow two-tower retrieval + ranker, FAISS serving, offline A/B over model variants.
Two-tower reading recommender plus an off-policy 'opportunity lab' — retrieval, LambdaMART reranking and causal evaluation.
Two-tower retrieval recommender on MovieLens with FAISS HNSW serving and diversity reranking.
PyTorch two-tower (dual-encoder) retrieval recommender with in-batch negatives and Recall@K/NDCG evaluation.
A three-stage recommender under a 50 ms deadline: two-tower retrieval, HNSW search, LambdaMART reranking. Measured p99 of 20.5 ms at 50 req/s, sheds quality instead of timing out under load, catches a corrupted index in one canary cycle, corrects position bias in clicks. CPU-only, offline.
To associate your repository with the two-tower topic, visit your repo's landing page and select "manage topics."