Ignacy Stepka
@igstepka
Followers
24
Following
211
Media
4
Statuses
20
PhD student @mldcmu
Pittsburgh, PA
Joined September 2017
RT @inverse_hessian: cool stuff to see at @kdd_news, including results from our senior thesis! #KDD2025.
0
1
0
The most fun image & video creation tool in the world is here. Try it for free in the Grok App.
0
58
506
@LangoMateusz @PUT_Poznan @AutonLab @CMU_Robotics 📅 Tuesday 5:45 pm - 8:00 pm in Exhibit Hall F poster no. 437. @LSztukiewicz will present our joint work on the relationship between saliency maps and fairness as part of the Undergraduate and Master’s Consortium. @PUT_Poznan @inverse_hessian. 📄 Paper:
arxiv.org
The widespread adoption of machine learning systems has raised critical concerns about fairness and bias, making mitigating harmful biases essential for AI development. In this paper, we...
1
0
0
@LangoMateusz @PUT_Poznan 📅 Monday 8:00 am - 12:00 pm in Room 700. Presenting our work on mitigating persistent client dropout in decentralized federated learning as part of the FedKDD workshop. @AutonLab @CMU_Robotics. 🌐 Project website: 📄 Paper:
1
0
1
📅 Tuesday 5:30 - 8 pm (poster no. 141) and Friday 8:55 - 9:15 (Room 801 A, talk). I’ll be giving a talk and presenting a poster on robust counterfactual explanations. @LangoMateusz @PUT_Poznan. 🌐 Project website: 📄 Paper:
arxiv.org
Counterfactual explanations (CFEs) guide users on how to adjust inputs to machine learning models to achieve desired outputs. While existing research primarily addresses static scenarios,...
1
0
1
RT @inverse_hessian: Curious about what Time Series Foundation Models actually learn? Stop by our poster today at #ICML2025! .Presented by….
0
3
0
RT @MononitoGoswami: Wednesday, 4:30 PM PDT. Presenting our work Exploring Representations and Interventions in Time Series Foundation Mode….
arxiv.org
Time series foundation models (TSFMs) promise to be powerful tools for a wide range of applications. However, their internal representations and learned concepts are still not well understood. In...
0
1
0
RT @CHILconference: Excited to highlight @WPotosnak et al.'s work: a novel hybrid global-local architecture + model-agnostic pharmacokineti….
0
4
0
Explore more:. 📄 paper: 👨💻 code: 🌐 project page: 👏 Big thanks to my co-authors Jerzy Stefanowski and @LangoMateusz ! #KDD2025 #TrustworthyAI #XAI 7/7🧵.
github.com
Counterfactual Explanations with Probabilistic Guarantees on their Robustness to Model Change (KDD'25) - istepka/betarce
0
0
1
📊 Results: Across 6 datasets, BetaRCE consistently achieved target robustness levels while preserving explanation quality and maintaining a competitive robustness-cost trade-off. 6/7🧵
1
0
0
You control both confidence level (α) and robustness threshold (δ), giving mathematical guarantees your explanation will survive changes! For formal proofs on optimal SAM sampling methods and the full theoretical foundation, check out our paper! 5/7🧵
1
0
0
⚙️ Under the hood: BetaRCE explores a "Space of Admissible Models" (SAM) - representing expected/foreseeable changes to your model. Using Bayesian statistics, we efficiently estimate the probability that explanations remain valid across these changes. 4/7🧵.
1
0
0
✅ Our solution: BetaRCE - offers probabilistic guarantees for robustness to model change. It works with ANY model class, is post-hoc, and can enhance your current counterfactual methods. Plus, it allows you to control the robustness-cost trade-off. 3/7🧵
1
0
0
❌ This happens constantly in real-world AI systems. Current explanation methods don't address this well - they're limited to specific models, require extensive tuning, or lack guarantees about explanation robustness. 2/7🧵.
1
0
0
📣 New paper at #KDD2025 on robust counterfactual explanations!. Imagine an AI tells you "Increase income by $200 to get a loan". You do it, but when you reapply, the model has been updated and rejects you anyway. We solve this issue by making CFEs robust to model changes! 1/7🧵
1
4
8
RT @ghostdayamlc: We’re thrilled to announce the first keynote speaker for GHOST DAY: Applied Machine Learning Conference 2025 . Prof. Dr.….
0
5
0
RT @CMU_Robotics: The RI Summer Scholars showcased their research yesterday with a fantastic turnout! 46 undergrads from 11 different count….
0
3
0