Seminars

---

Title: Xai Library -  tutorial

When: 14/12/13

Where: Officine Garibaldi.

Abstract: The Kode team (Paolo Cintia and Andrea Spinelli) will present the recent refactoring of the XAI Library to the XAI group, offering a tutorial in which they will illustrate the use of the new library.

Streaming: The link to follow the online tutorial will be available on the Events channel in the Teams Group: XAI@KDD. 

----------

Microsoft Teams meetingJoin on your computer, mobile app or room deviceClick here to join the meeting<https://teams.microsoft.com/l/meetup-join/19%3a511689e6d6494b2c95e95fe823c57aae%40thread.tacv2/1702308391444?context=%7b%22Tid%22%3a%22c7456b31-a220-47f5-be52-473828670aa1%22%2c%22Oid%22%3a%22729b4d16-0567-46a8-a742-d2ae1bf09a4a%22%7d>Meeting ID: 362 749 499 63Passcode: 7bBvuQ 

---

When: 30/11/2023 at 11:30

Where: Officine Garibaldi. Stream details below.

Abstract: While existing clustering methods only provide the assignment of records to clusters without justifying the partitioning, we propose tree-based clustering methods that offer interpretable data partitioning through a shallow decision tree. These decision trees enable easy-to-understand explanations of cluster assignments through short and understandable split conditions. The proposed methods are evaluated through experiments on synthetic and real datasets and proved to be more effective than traditional clustering approaches and interpretable ones in terms of standard evaluation measures and runtime.

 

 

 

_____________

Microsoft Teams meeting

Join on your computer, mobile app or room device

Click here to join the meeting

Meeting ID: 366 914 674 820 

 

Passcode: 9gKaZe 

Download Teams | Join on the web

Or call in (audio only)

+39 02 3056 4191,,906266144#   Italy, Milano

Phone Conference ID: 906 266 144# 

Find a local number | Reset PIN

Learn More | Meeting options

 
 

_____________

---

In this talk, Laura will present us her work on ReasonX, a CLP-based tool for interactive reasoning-based explainability.

 

When & Where. Thursday 12th, 11:30 @ Officine Garibaldi. Stream details below.

Abstract. Explaining opaque Machine Learning (ML) models is an increasingly relevant problem. Current explanation in AI (XAI) methods suffer several shortcomings, among others an insufficient incorporation of background knowledge, and a lack of abstraction and interactivity with the user. We propose REASONX, an explanation method based on Constraint Logic Programming (CLP). REASONX can provide declarative, interactive explanations for decision trees, which can be the ML models under analysis or global/local surrogate models of any black-box model. Users can express background or common sense knowledge using linear constraints and MILP optimization over features of factual and contrastive instances, and interact with the answer constraints at different levels of abstraction through constraint projection. We present here the architecture of REASONX, which consists of a Python layer, closer to the user, and a CLP layer. REASONX's core execution engine is a Prolog meta-program with declarative semantics in terms of logic theories.

____

Microsoft Teams meeting

Join on your computer, mobile app or room device

Click here to join the meeting

Meeting ID: 341 121 035 466 

 

Passcode: BHbWct 

---

Welcome seminar by new postdoc@UniPI, Antonio Mastropietro, who will tell us about his past and current research interests.

When & Where October 5th, 11:30am. Sala Polifunzionale @ CS department.
Stream details below.

_____

Microsoft Teams meeting

Join on your computer, mobile app or room device

Click here to join the meeting

Meeting ID: 361 846 547 698 

 

Passcode: wyaqvW 

---

Dear KDDers, in this talk Jose will present us his FAccT 2023 papere on Domain Adaptive Decision Trees, feel free to join!

 

Title. Domain Adaptive Decision Trees: Implications for Accuracy and Fairness

Abstract. In uses of pre-trained machine learning models, it is a known issue that the target population in which the model is being deployed may not have been reflected in the source population with which the model was trained. This can result in a biased model when deployed, leading to a reduction in model performance. One risk is that, as the population changes, certain demographic groups will be under-served or otherwise disadvantaged by the model, even as they become more represented in the target population. The field of domain adaptation proposes techniques for a situation where label data for the target population does not exist, but some information about the target distribution does exist. In this paper we contribute to the domain adaptation literature by introducing domain-adaptive decision trees (DADT). We focus on decision trees given their growing popularity due to their interpretability and performance relative to other more complex models. With DADT we aim to improve the accuracy of models trained in a source domain (or training data) that differs from the target domain (or test data). We propose an in-processing step that adjusts the information gain split criterion with outside information corresponding to the distribution of the target population. We demonstrate DADT on real data and find that it improves accuracy over a standard decision tree when testing in a shifted target population. We also study the change in fairness under demographic parity and equal opportunity. Results show an improvement in fairness with the use of DADT.

 When & Where. June 1st, 2pm. Sala Polifunzionale @ Department of Computer Science. Stream available here.

Topics. Decision Trees, Fairness, Explainability

---

Title: Explanation visualization at scale: progress on decision rules and feature importance

Abstract: Starting from a more user-centered approach to XAI our latest work proposes a new visualization method for the combination of rules and features importance for datasets with a large number of features. We will be showing an interface through which the users can interact with these visualizations.

____

Microsoft Teams meeting

Join on your computer, mobile app or room deviceClick here to join the meeting

Meeting ID: 372 276 263 699
Passcode: Q9Vnq2
Download Teams | Join on the web

Or call in (audio only)
+39 02 3056 4191,,648642497#   Italy, MilanoPhone Conference ID: 648 642 497#
Find a local number | Reset PIN
Learn More | Meeting options
________________________________________________________________________________

---

Abstract

Artificial Intelligence (AI) has come to prominence as one of the major components of our society, with applications in most aspects of our lives. In this field, complex and highly nonlinear machine learning models such as ensemble models, deep neural networks, and Support Vector Machines have consistently shown remarkable accuracy in solving complex tasks. Although accurate, AI models often are “black boxes” which we are not able to understand. Relying on these models has a multifaceted impact and raises significant concerns about their transparency. Applications in sensitive and critical domains are a strong motivational factor in trying to understand the behavior of black boxes. We propose to address this issue by providing an interpretable layer on top of black box models by aggregating “local” explanations. We present GLocalX, a “local-first” model agnostic explanation method. Starting from local explanations expressed in form of local decision rules, GLocalX iteratively generalizes them into global explanations by hierarchically aggregating them. Our goal is to learn accurate yet simple interpretable models to emulate the given black box, and, if possible, replace it entirely. We validate GLocalX in a set of experiments in standard and constrained settings with limited or no access to either data or local explanations. Experiments show that GLocalX is able to accurately emulate several models with simple and small models, reaching state-of-the-art performance against natively global solutions. Our findings show how it is often possible to achieve a high level of both accuracy and comprehensibility of classification models, even in complex domains with high-dimensional data, without necessarily trading one property for the other. This is a key requirement for a trustworthy AI, necessary for adoption in high-stakes decision making applications.

___

Microsoft Teams meeting

Join on your computer, mobile app or room deviceClick here to join the meeting

Meeting ID: 340 547 739 918
Passcode: FNLU3Y
Download Teams | Join on the web

Or call in (audio only)
+39 02 3056 4191,,780108878#   Italy, MilanoPhone Conference ID: 780 108 878#
Find a local number | Reset PIN
Learn More | Meeting options

---

Abstract: Search engines decide what we see for a given search query. Since many people are exposed to information through search engines, it is fair to expect that search engines are neutral. However, search engine results do not necessarily cover all the viewpoints of a search query topic, and they can be biased towards a specific view since search engine results are returned based on relevance, which is calculated using many features and sophisticated algorithms where search neutrality is not necessarily the focal point. Therefore, it is important to evaluate the search engine results with respect to bias. In this seminar, we will firstly examine the stance (in support or against), as well as the ideological bias (conservative or liberal) in search results of two popular search engines with respect to controversial query topics such as abortion, medical marijuana, and gay marriage.

In the second part of this seminar, we will investigate gender bias in online education. Students are increasingly using online materials to learn new subjects or to supplement their learning process in educational institutions. Nonetheless, online educational materials in terms of possible gender bias and stereotypes which may appear in different forms are yet to be investigated in the context of search bias in a widely-used search platform. Thus, as a first step towards measuring possible gender bias, we will analyse online educational videos of YouTube in terms of the perceived, i.e. from the viewer's perspective, gender of their narrators. Then, we will evaluate possible perceived gender bias in ranked educational video search results returned by YouTube in response to queries related to STEM (Science, Technology, Engineering, and Mathematics) and NON-STEM fields of education. Lastly, we will also discuss our attempts for bias mitigation in the scope of perceived gender bias in YouTube.

---

Microsoft Teams meeting

Join on your computer, mobile app or room deviceClick here to join the meeting

Meeting ID: 322 172 125 021
Passcode: re9bSM
Download Teams | Join on the web

Or call in (audio only)
+39 02 3056 4191,,261413320#   Italy, Milano Phone Conference ID: 261 413 320#
Find a local number | Reset PIN
Learn More | Meeting options