SEA: Learning to Rank

This SEA will be all about ranking. We have two amazing speakers lined up: Thorsten Joachims from Cornell University and Raviteja Anantha from Apple.

*** IMPORTANT: Make sure to (1) attend the meetup on the meetup page and (2) ensure you receive emails from Meetup. Shortly before the event we will send you the Zoom link and password to attend, as well as the info you need to log in via the browser (if your organisation does not allow you to install Zoom). You will only receive this if you have done both these steps. ***

** 17:00 - 17:30 CET - Thorsten Joachims, Cornell University **

Title: Fair Ranking with Biased Data

Abstract: Search engines and recommender systems have become the dominant matchmaker for a wide range of human endeavors -- from online retail to finding romantic partners. Consequently, they carry immense power in shaping markets and allocating opportunity to the participants. In this talk, I will discuss how the machine learning algorithms underlying these system can produce unfair ranking policies for both exogenous and endogenous reasons. Exogenous reasons often manifest themselves as biases in the training data, which then get reflected in the learned ranking policy and lead to rich-get-richer dynamics. But even when trained with unbiased data, reasons endogenous to the algorithms can lead to unfair or undesirable allocation of opportunity. To overcome these challenges, I will present new machine learning algorithms that directly address both endogenous and exogenous unfairness.

** 17:30 - 18:00 CET - Raviteja Anantha, Apple **

Title: Learning to Rank Intents in Voice Assistants

Abstract: Voice Assistants aim to fulfill user requests by choosing the best intent from multiple options generated by its Automated Speech Recognition and Natural Language Understanding sub-systems. However, voice assistants do not always produce the expected results. This can happen because voice assistants choose from ambiguous intents—user-specific or domain-specific contextual information reduces the ambiguity of the user request. Additionally the user information-state can be leveraged to understand how relevant/executable a specific intent is for a user request. In this work, we propose a novel Energy-based model for the intent ranking task, where we learn an affinity metric and model the trade-off between extracted meaning from speech utterances and relevance/executability aspects of the intent. Furthermore we present a Multisource Denoising Autoencoder based pretraining that is capable of learning fused representations of data from multiple sources. We empirically show our approach outperforms existing state of the art methods by reducing the error-rate by 3.8%, which in turn reduces ambiguity and eliminates undesired dead-ends leading to better user experience. Finally, we evaluate the robustness of our algorithm on the intent ranking task and show our algorithm improves the robustness by 33.3%.

After the two talks we will leave the Zoom call open for another half an hour, for any remaining questions. This always results in a nice discussion!