Houssam Nassif, PhD

Welcome to my research page. Until 12/2022, I was a Principal Applied Scientist at Amazon. I am currently at Meta. For a glimpse at my projects and what I do, check these blog posts and products:

• The science behind Amazon Prime blog post
  
• What's new in recommender systems blog post
  
• A/B Testing at Scale blog post and summary video
  
• The morning paper blog post
  
• Shop-by-look interactive visual browsing platform and impact
  
• Diverse visual stream experience and news article
  
• Amazon Instant Video recommendations
  
• Amazon Prime recommendations and impact
  
• Machine Learning Scientist role video
  

Journals
• Bayesian Meta-Prior Learning Using Empirical Bayes
  Nabi S, Nassif H, Hong J, Mamani H, and Imbens G
  Management Science, 68(3):1737-1755, 2022.
  (Paper, arXiv paper version, Slides, Video)
  
  
• Rational Polypharmacology: Systematically Identifying and Engaging Multiple Drug Targets To Promote Axon Growth (Cover Article)
  Al-Ali H, Lee DH, Danzi MC, Nassif H, Gautam P, Wennerberg K, Zuercher B, Drewry DH, Lee JK, Lemmon VP, and Bixby JL
  ACS Chemical Biology, 10(8):1939-1951, 2015.
  (Paper, Cover graphic, Follow-up Poster, Follow-up Poster)
  
  
• Predicting Invasive Breast Cancer Versus DCIS in Different Age Groups
  Ayvaci M, Alagoz O, Chhatwal J, Munoz del Rio A, Sickles EA, Nassif H, Kerlikowske K, and Burnside ES
  BMC Cancer, 14:584, 2014.
  (Paper)
  
  
• Automatic Classification of Mammography Reports by BI-RADS Breast Tissue Composition Class
  Percha B, Nassif H, Lipson J, Burnside E, and Rubin D
  Journal of the American Medical Informatics Association (JAMIA), 19(5):913-916, 2012.
  (Paper, Code)
  
  
• Automated Identification of Protein-Ligand Interaction Features Using Inductive Logic Programming: A Hexose Binding Case Study
  Santos JCA, Nassif H, Page D, Muggleton SH, and Sternberg MJE
  BMC Bioinformatics, 13:162, 2012.
  (Paper, Code, Data, Input)
  
  
• Prediction of Protein-Glucose Binding Sites Using Support Vector Machines
  Nassif H, Al-Ali H, Khuri S, and Keyrouz W
  Proteins: Structure, Function and Bioinformatics, 77(1):121-132, 2009.
  (Paper, Code, Data, Input)
  
  
Conferences
• Best of Three Worlds: Adaptive Experimentation for Digital Marketing in Practice
  Fiez T, Nassif H, Chen YC, Gamez S, Jain L
  The Web Conference (WWW'24), Singapore, 2024.
  (Paper, Summary video, Poster)
  
  
• Experimental Designs for Heteroskedastic Variance
  Weltz J, Fiez T, Laber E, Volfovsky A, Mason B, Nassif H, and Jain L
  Conference on Neural Information Processing Systems (NeurIPS'23), New Orleans, 2023.
  (Paper, Poster)
  
  
• A Data-Driven State Aggregation Approach for Dynamic Discrete Choice Models
  Geng S, Nassif H, and Manzanares CA
  The Conference on Uncertainty in Artificial Intelligence (UAI'23), Pittsburgh, PA, pp. 647-657, 2023.
  (Paper, Poster)
  
  
• Neural Insights for Digital Marketing Content Design
  Kong F, Li Y, Nassif H, Fiez T, Chakrabarti S, and Henao R
  International Conference on Knowledge Discovery and Data Mining (KDD'23), Long Beach, CA, pp. 4320-4332, 2023.
  (Paper, Poster, Summary video, Slides)
  
  
• Enhancing Transistor Sizing in Analog IC Design using a Circuit-Focused Semi-Supervised Learning
  Mina R, Sakr GE, and Nassif H
  IEEE International Multidisciplinary Conference on Engineering Technology (IMCET'23), Beirut, Lebanon, pp. 223-228, 2023.
  (Paper, Slides)
  
  
• Instance-Optimal PAC Algorithms for Contextual Bandits
  Li Z, Ratliff LJ, Nassif H, Jamieson K, and Jain L
  Conference on Neural Information Processing Systems (NeurIPS'22), New Orleans, pp. 37590-37603, 2022.
  (Paper, Slides, Poster, Video)
  
  
• Adaptive Experimental Design and Counterfactual Inference
  Fiez T, Gamez S, Chen A, Nassif H, and Jain L
  ACM Conference on Recommender Systems (RecSys'22) Workshops, Seattle, 2022.
  (Paper, Slides, Poster, Video)
  
  
• Improved Confidence Bounds for the Linear Logistic Model and Applications to Bandits
  Jun KS, Jain L, Mason B, and Nassif H
  International Conference on Machine Learning (ICML'21), Virtual, pp. 5148-5157, 2021.
  (Paper, Slides, Extended slides, Video)
  
  
• Deep PQR: Solving Inverse Reinforcement Learning using Anchor Actions
  Geng S, Nassif H, Manzanares C, Reppen M, and Sircar R
  International Conference on Machine Learning (ICML'20), Virtual, pp. 3431-3441, 2020.
  (Paper, Slides, Additional slides, Video, Additional video)
  
  
• Seeker: Real-Time Interactive Search
  Biswas A, Pham TT, Vogelsong M, Snyder B, and Nassif H
  International Conference on Knowledge Discovery and Data Mining (KDD'19), Anchorage, Alaska, pp. 2867-2875, 2019.
  (Paper, Poster, Summary video, Slides, Video)
  
  
• Contextual Multi-Armed Bandits for Causal Marketing (Amazon Research Scientist Summit Best Paper Award)
  Sawant N, Namballa CB, Sadagopan N, and Nassif H
  International Conference on Machine Learning (ICML'18) Workshops, Stockholm, Sweden, 2018.
  (Paper, Poster)
  
  
• An Efficient Bandit Algorithm for Realtime Multivariate Optimization (Audience Appreciation Award)
  Hill D, Nassif H, Liu Y, Iyer A, and Vishwanathan SVN
  International Conference on Knowledge Discovery and Data Mining (KDD'17), Halifax, Canada, pp. 1813-1821, 2017.
  (Paper, Summary video, Poster, Additional slides, Award)
  
  
• Adaptive, Personalized Diversity for Visual Discovery (Best Short Paper Award)
  Teo CH, Nassif H, Hill D, Srinavasan S, Goodman M, Mohan V, and Vishwanathan SVN
  ACM Conference on Recommender Systems (RecSys'16), Boston, pp. 35-38, 2016.
  (Paper, Slides, Video, Additional slides, Award)
  
  
• Diversifying Music Recommendations
  Nassif H, Cansizlar KO, Goodman M, and Vishwanathan SVN
  International Conference on Machine Learning (ICML'16) Workshops, New York, 2016.
  (Paper, Slides, Poster)
  
  
• Support Vector Machines for Differential Prediction
  Kuusisto F, Santos Costa V, Nassif H, Burnside ES, Page D, and Shavlik J
  European Conference on Machine Learning (ECML'14), Nancy, France, pp. 50-65, 2014.
  (Paper, Slides, Poster, Code)
  
  
• Score As You Lift (SAYL): A Statistical Relational Learning Approach to Uplift Modeling
  Nassif H, Kuusisto F, Burnside ES, Page D, Shavlik J, and Santos Costa V
  European Conference on Machine Learning (ECML'13), Prague, pp. 595-611, 2013.
  (Paper, Slides, Poster, Code)
  
  
• Uplift Modeling with ROC: An SRL Case Study
  Nassif H, Kuusisto F, Burnside ES, and Shavlik J
  International Conference on Inductive Logic Programming (ILP'13), Rio de Janeiro, Brazil, pp. 40-45, 2013.
  (Paper, Slides, Poster)
  
  
• Genetic Variants Improve Breast Cancer Risk Prediction on Mammograms
  Liu J, Page D, Nassif H, Shavlik J, Peissig P, McCarty C, Onitilo AA, and Burnside E
  American Medical Informatics Association Symposium (AMIA'13), Washington, DC, pp. 876-885, 2013.
  (Paper, Slides)
  
  
• Using Machine Learning to Identify Benign Cases with Non-Definitive Biopsy
  Kuusisto F, Dutra I, Nassif H, Wu Y, Klein M, Neuman H, Shavlik J, and Burnside E
  IEEE International Conference on E-Health Networking, Application & Services (HealthCom'13), Lisbon, Portugal, pp. 283-285, 2013.
  (Paper, Poster)
  
  
• Relational Differential Prediction
  Nassif H, Santos Costa V, Burnside ES, and Page D
  European Conference on Machine Learning (ECML'12), Bristol, UK, pp. 617-632, 2012.
  (Paper, Slides, Poster, Code, Michalski trains data)
  
  
• Logical Differential Prediction Bayes Net, Improving Breast Cancer Diagnosis for Older Women
  Nassif H, Wu Y, Page D, and Burnside ES
  American Medical Informatics Association Symposium (AMIA'12), Chicago, pp. 1330-1339, 2012.
  (Paper, Slides, Logic code)
  
  
• Extracting BI-RADS Features from Portuguese Clinical Texts
  Nassif H, Cunha F, Moreira IC, Cruz-Correia R, Sousa E, Page D, Burnside E, and Dutra I
  IEEE International Conference on Bioinformatics and Biomedicine (BIBM'12), Philadelphia, pp. 539-542, 2012.
  (Paper, Slides, Code)
  
  
• Integrating Machine Learning and Physician Knowledge to Improve the Accuracy of Breast Biopsy
  Dutra I, Nassif H, Page D, Shavlik J, Strigel RM, Wu Y, Elezaby ME, and Burnside E
  American Medical Informatics Association Symposium (AMIA'11), Washington, DC, pp. 349-355, 2011.
  (Paper, Slides)
  
  
• Uncovering Age-Specific Invasive and DCIS Breast Cancer Rules Using Inductive Logic Programming
  Nassif H, Page D, Ayvaci M, Shavlik J, and Burnside ES
  ACM International Health Informatics Symposium (IHI'10), Arlington, VA, USA, pp. 76-82, 2010.
  (Paper, Slides, Code)
  
  
• Information Extraction for Clinical Data Mining: A Mammography Case Study
  Nassif H, Woods R, Burnside ES, Ayvaci M, Shavlik J, and Page D
  IEEE International Conference on Data Mining (ICDM'09) Workshops, Miami, pp. 37-42, 2009.
  (Paper, Slides, Code)
  
  
• An Inductive Logic Programming Approach to Validate Hexose Binding Biochemical Knowledge
  Nassif H, Al-Ali H, Khuri S, Keyrouz W, and Page D
  International Conference on Inductive Logic Programming (ILP'09), Leuven, Belgium, pp. 149-165, 2009.
  (Paper, Slides, Feature extraction code, ILP code, Data, Input)
  
  
Patents
• Artificial Intelligence System For Optimizing Network-Accessible Content
  Nassif H, Hill D, Hu T, Iyer AM, Liu J, Liu Y, Srinivasan S, and Swaminathan V
  US Patent 11126785, 2021.
  (Patent)
  
  
• Content Selection Algorithms
  Srinavasan S, Nassif H, Mohan V, Swaminathan V, and Goodman MH
  US Patent 9817846, 2017.
  (Patent)
  
  
Theses
• Differential Relational Learning
  Nassif H
  PhD Dissertation, University of Wisconsin - Madison, 2012.
  (Thesis, Slides)
  
  
• A Pattern Recognition Based Model for Characterizing and Predicting Glucose-Binding Sites
  Nassif H
  MS Thesis, American University of Beirut, 2006.
  (Thesis, Slides)