Interpretability is crucial for doctors, hospitals, pharmaceutical companies and biotechnology corporations to analyze and make decisions for high stakes problems that involve human health. Tree-based methods have been widely adopted for survival analysis due to their appealing interpretablility and their ability to capture complex relationships. However, most existing methods to produce survival trees rely on heuristic (or greedy) algorithms, which risk producing sub-optimal models. We present a dynamic-programming-with-bounds approach that finds provably-optimal sparse survival tree models, frequently in only a few seconds.

Regression trees are one of the oldest forms of AI models, and their predictions can be made without a calculator, which makes them broadly useful, particularly for high-stakes applications. Within the large literature on regression trees, there has been little effort towards full provable optimization, mainly due to the computational hardness of the problem. This work proposes a dynamic-programming-with-bounds approach to the construction of provably-optimal sparse regression trees. We leverage a novel lower bound based on an optimal solution to the k-Means clustering algorithm on one dimensional data. We are often able to find optimal sparse trees in seconds, even for challenging datasets that involve large numbers of samples and highly-correlated features.

In today's web ecosystem, a website that uses a Content Delivery Network (CDN) shares its Transport Layer Security (TLS) private key or session key with the CDN. In this paper, we present the design and implementation of InviCloak, a system that protects the confidentiality and integrity of a user and a website's private communications without changing TLS or upgrading a CDN. InviCloak builds a lightweight but secure and practical key distribution mechanism using the existing DNS infrastructure to distribute a new public key associated with a website's domain name. A web client and a website can use the new key pair to build an encryption channel inside TLS. InviCloak accommodates the current web ecosystem. A website can deploy InviCloak unilaterally without a client's involvement to prevent a passive attacker inside a CDN from eavesdropping on their communications. If a client also installs InviCloak's browser extension, the client and the website can achieve end-to-end confidential and untampered communications in the presence of an active attacker inside a CDN. Our evaluation shows that InviCloak increases the median page load times (PLTs) of realistic web pages from 2.0s to 2.1s, which is smaller than the median PLTs (2.8s) of a state-of-the-art TEE-based solution.

In any given machine learning problem, there may be many models that could explain the data almost equally well. However, most learning algorithms return only one of these models, leaving practitioners with no practical way to explore alternative models that might have desirable properties beyond what could be expressed within a loss function. The Rashomon set is the set of these all almost-optimal models. Rashomon sets can be extremely complicated, particularly for highly nonlinear function classes that allow complex interaction terms, such as decision trees. We provide the first technique for completely enumerating the Rashomon set for sparse decision trees; in fact, our work provides the first complete enumeration of any Rashomon set for a non-trivial problem with a highly nonlinear discrete function class. This allows the user an unprecedented level of control over model choice among all models that are approximately equally good. We represent the Rashomon set in a specialized data structure that supports efficient querying and sampling. We show three applications of the Rashomon set: 1) it can be used to study variable importance for the set of almost-optimal trees (as opposed to a single tree), 2) the Rashomon set for accuracy enables enumeration of the Rashomon sets for balanced accuracy and F1-score, and 3) the Rashomon set for a full dataset can be used to produce Rashomon sets constructed with only subsets of the data set. Thus, we are able to examine Rashomon sets across problems with a new lens, enabling users to choose models rather than be at the mercy of an algorithm that produces only a single model.