I typically run compute jobs remotely using my M1 Macbook as a terminal. So, when PyTorch recently launched its backend compatibility with Metal on M1 chips, I was kind of interested to see what kind of GPU acceleration performance can be achieved. To make the process super easy, Anaconda also recently released an M1-native version. … Continue reading PyTorch GPU acceleration on M1 Mac
If you follow the news on deep learning, you may have encountered the applications of Nvidia’s StyleGAN that generates photo-realistic fake faces. Since the typical dataset that StyleGAN is trained on consist of amateur-produced portraits from Flickr, I was curious to see what StyleGAN would learn from social media posts. I crawled over 1M social … Continue reading AdStyleGAN
Happy to share the pre-publication version of this paper just accepted at Marketing Science entitled “How support for Black Lives Matter impacts consumer responses on social media” coauthored with Marco Qin, Xueming Luo, and Eric (Yu) Kou (PhD student) In this paper, we show that online support for the BLM movement is associated with decreased follower … Continue reading Social media receptivity of brands’ BLM support
Here’s a set of notes on regression adjustments and propensity score and coarsened exact matching that I created for a training session at Amazon Ads. The code is in Python mostly for pedagogical reasons, should really just use the Matchit library in R. Continue reading Matching notes
Working on making a repository of notebooks demonstrating applications of AI/ML to business problems with real datasets. This is the first installment of this series. In this notebook, I cover Neural network architecture Embedding layers LSTM Layers Attention layers Training details and model diagnostics The application is in predicting multi-dimensional hotel quality ratings (service, rooms, … Continue reading (Code) Deep Learning for NLP: Predicting multi-dimensional hotel ratings
Many platforms employ a mix of algorithms and humans for recommending products. For example, Spotify features algorithm and editor / celebrity playlists. Amazon employs both algorithmic product recommendations and “staff picks.” Apple News features experienced editors in conjunction with algorithms when recommending products. This phenomenon begs the question, why do these big platforms, run by … Continue reading Do consumers prefer AI or human to recommend products?
The bullet point takeaways: Track crowds. Economic devastation happened with or without SIP. Link to most recent draft of working paper Like many academics, I’ve recently tossed my proverbial thinking hat into the ring to investigate the causal impact of shelter in place (SIP) policies on public health and social distancing outcomes. This is clearly … Continue reading Did Shelter in Place policies work?
First HBR now online! Much thanks to coauthors KT Manis and Alex Chaudhry for putting this concise summary of our previous JMR together. https://hbr.org/2020/05/5-principles-for-responding-to-customer-reviews The takeaways from the takeaways: Principle 1. Address a positive online review by providing a generic, short response. Principle 2. Delay responses for positive reviews. Principle 3. Respond to all negative … Continue reading 5 principles for responding to customer reviews
Wholefoods Getting groceries delivered has been a bit of a challenge recently. I had been getting deliveries from Reading Market (through Mercado which subcontracts DoorDash(?)). Their lead times are now over 2 weeks. It appears WholeFoods is taking a different approach by releasing delivery slots at random times throughout the day. However, every time I … Continue reading Sniping that WholeFoods delivery slot and other web scraping projects
If you’re anything like me, you avoid calling customer service at all costs. The anticipated pain, whether rational or not, of waiting on hold and having to deal with another human interaction is sometimes just too much of a barrier to making a phone call. I often end up using automated chat functions even though … Continue reading Designing an Effective Service Chatbot by Anticipating Customer Needs