Chu Wang

Manager of Applied Science, Amazon Ads

I’m an applied machine learning and applied mathematics practitioner

About Me

I'm interested in ML/Math and their applications which impact science communities, industries, and beyond. Currently, I'm working on NLU, personalization, and reinforcement learning for sponsored ads. Previously, I have worked on network science, ensemble methods, large-scale simulations, and computational mathematics.


To contact me:

chuwang [THE-AT-SIGN] amazon.com (for Amazon related work)

chuwang.math [THE-SAME-SIGN] gmail.com (for general academic work)

About My work

MAB with Binary Feedbacks

A knowledge gradient policy with provable bounds for sequential decision making with stochastic binary feedbacks.

Iterated Learning

Repeated Bayesian learning sessions will inevitably collapse to priors unless reinforced with logarithmic-size sessions.

ProductNet

Clean small data is better than noisy big data for product representation learning via multimodal deep models.

Whole Page Recommendation

Whole-page optimization via ordered combinatorial semi-bandits and network flow is applied to news recommendation.

Reference Product Search

Billion-scale product search via GPU-based approximate KNN in product vector space constructed via attention auto-encoder.

Fractional Quantum Mechanics

Scalable and robust numerical methods let you know how wave function evolves and propagates in a fractal universe.

Polymer Self Assembly

Large-scale simulations on diblock-polymer self-organize into complex ordered metastable structures.

Pack-Size Price Consistency

Pack-size and multi-pack info are inferred from product metadata to ensure consistent unit price for different products.

Regularized Boosting

A Frank-Wolfe type regularized boosting algorithm with provable generalization bounds via Rademaker complexity.

Polymer Interfacial Structures

The dynamics of stable phase eating metastable phase are simulated as interface moving during phase transition.

Graph Diffusion Quasi-Metrics

The similarity between nodes are modeled as Markov transition probability. This similarity is proved to be a quasi-metric.

Epitaxies between Structures

Ordered phases prefer certain orientations when growing from metastable structures. There could be multiple preferred ones.