Zobrazeno 1 - 10
of 22
pro vyhledávání: '"Josif Grabocka"'
Publikováno v:
SIGIR
Normalized discounted cumulative gain (NDCG) is one of the popular evaluation metrics for recommender systems and learning-to-rank problems. As it is non-differentiable, it cannot be optimized by gradient-based optimization procedures. In the last tw
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::30eaa05aabb810003e6a7a7e161a0808
https://doi.org/10.1145/3404835.3462864
https://doi.org/10.1145/3404835.3462864
Autor:
Michael Ruchte, Josif Grabocka
Multi-objective optimization (MOO) is a prevalent challenge for Deep Learning, however, there exists no scalable MOO solution for truly deep neural networks. Prior work either demand optimizing a new network for every point on the Pareto front, or in
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::2be6e81d95a490475b6e05e02ebce1ca
http://arxiv.org/abs/2103.13392
http://arxiv.org/abs/2103.13392
Publikováno v:
Machine Learning and Knowledge Discovery in Databases. Research Track ISBN: 9783030864859
ECML/PKDD (1)
ECML/PKDD (1)
The computational challenges arising from increasingly large search spaces in hyperparameter optimization necessitate the use of performance prediction methods. Previous works have shown that approximated performances at various levels of fidelities
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::69a0a3bff9e662d1cc14665935b79cf1
https://doi.org/10.1007/978-3-030-86486-6_30
https://doi.org/10.1007/978-3-030-86486-6_30
Publikováno v:
Advances in Knowledge Discovery and Data Mining ISBN: 9783030474256
PAKDD (1)
PAKDD (1)
Self-supervised learning is a promising new technique for learning representative features in the absence of manual annotations. It is particularly efficient in cases where labeling the training data is expensive and tedious, naturally linking it to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::665ee59c98c8998f6d6e697eaad45caa
https://doi.org/10.1007/978-3-030-47426-3_39
https://doi.org/10.1007/978-3-030-47426-3_39
Publikováno v:
RecSys
In very sparse recommender data sets, attributes of users such as age, gender and home location and attributes of items such as, in the case of movies, genre, release year, and director can improve the recommendation accuracy, especially for users an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::05dd2980e2e8d138499dab840c4790e0
https://doi.org/10.1145/3298689.3346999
https://doi.org/10.1145/3298689.3346999
Publikováno v:
KDD
The task of classifying multi-relational data spans a wide range of domains such as document classification in citation networks, classification of emails, and protein labeling in proteins interaction graphs. Current state-of-the-art classification m
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a9bdc63f324251598e66a9baf7594ff6
https://doi.org/10.1145/3292500.3330863
https://doi.org/10.1145/3292500.3330863
Publikováno v:
IJCNN
Parking availability prediction is rapidly gaining interest within the community as an operationally cheap approach to identifying empty parking locations. Parking locations accommodate multiple vehicles and are rarely completely occupied. This makes
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::e66d5b6ff01a15ff60790ace48434804
https://doi.org/10.1109/ijcnn.2019.8852400
https://doi.org/10.1109/ijcnn.2019.8852400
Meta-learning, or learning to learn, is a machine learning approach that utilizes prior learning experiences to expedite the learning process on unseen tasks. As a data-driven approach, meta-learning requires meta-features that represent the primary
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6d0a0437374c03721ad8739b83de88a0
http://arxiv.org/abs/1905.11063
http://arxiv.org/abs/1905.11063
Publikováno v:
ICAART (2)
Multi-label network classification is a well-known task that is being used in a wide variety of web-based and non-web-based domains. It can be formalized as a multi-relational learning task for predicting nodes labels based on their relations within
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c3623c62dfae25c3a63f26cf85536c36