Zobrazeno 1 - 10
of 11 494
pro vyhledávání: '"Jeremias, A"'
As complex machine learning models continue to find applications in high-stakes decision-making scenarios, it is crucial that we can explain and understand their predictions. Post-hoc explanation methods provide useful insights by identifying importa
Externí odkaz:
http://arxiv.org/abs/2409.20427
We study 2d $SU(N)$ QCD with an adjoint Dirac fermion. Assuming that the IR limit of the massless theory is captured by a WZW coset CFT, we show that this CFT can be decomposed into a sum of distinct CFTs, each representing a superselection sector (u
Externí odkaz:
http://arxiv.org/abs/2409.17989
We analyse the impact of using tempered likelihoods in the production of posterior predictions. Our findings reveal that once the sample size is at least moderately large and the temperature is not too small, then likelihood tempering has virtually n
Externí odkaz:
http://arxiv.org/abs/2408.08806
Autor:
Le, Yanfang, Pan, Rong, Newman, Peter, Blendin, Jeremias, Kabbani, Abdul, Jain, Vipin, Sivaramu, Raghava, Matus, Francis
Emerging artificial intelligence (AI) and machine learning (ML) workloads present new challenges of managing the collective communication used in distributed training across hundreds or even thousands of GPUs. This paper presents STrack, a novel hard
Externí odkaz:
http://arxiv.org/abs/2407.15266
Autor:
Traub, Jeremias, Bungert, Till J., Lüth, Carsten T., Baumgartner, Michael, Maier-Hein, Klaus H., Maier-Hein, Lena, Jaeger, Paul F
Selective Classification, wherein models can reject low-confidence predictions, promises reliable translation of machine-learning based classification systems to real-world scenarios such as clinical diagnostics. While current evaluation of these sys
Externí odkaz:
http://arxiv.org/abs/2407.01032
Autor:
Lenzi, Jeremias
Voting mechanisms are widely accepted and used methods for decentralized decision-making. Ensuring the acceptance of the voting mechanism's outcome is a crucial characteristic of robust voting systems. Consider this scenario: A group of individuals w
Externí odkaz:
http://arxiv.org/abs/2407.01844
Sparse auto-encoders are useful for extracting low-dimensional representations from high-dimensional data. However, their performance degrades sharply when the input noise at test time differs from the noise employed during training. This limitation
Externí odkaz:
http://arxiv.org/abs/2406.16052
Autor:
Zenk, Maximilian, Zimmerer, David, Isensee, Fabian, Traub, Jeremias, Norajitra, Tobias, Jäger, Paul F., Maier-Hein, Klaus
Semantic segmentation is an essential component of medical image analysis research, with recent deep learning algorithms offering out-of-the-box applicability across diverse datasets. Despite these advancements, segmentation failures remain a signifi
Externí odkaz:
http://arxiv.org/abs/2406.03323
Autor:
Epperlein, Jeremias, Schmieding, Scott
Given a homeomorphism $T \colon X \to X$ of a compact metric space $X$, the stabilized automorphism group $\textrm{Aut}^{\infty}(T)$ of the system $(X,T)$ is the group of self-homeomorphisms of $X$ which commute with some power of $T$. We study the q
Externí odkaz:
http://arxiv.org/abs/2405.20463
Autor:
Teneggi, Jacopo, Sulam, Jeremias
Recent works have extended notions of feature importance to \emph{semantic concepts} that are inherently interpretable to the users interacting with a black-box predictive model. Yet, precise statistical guarantees, such as false positive rate contro
Externí odkaz:
http://arxiv.org/abs/2405.19146