Zobrazeno 1 - 10
of 83
pro vyhledávání: '"Zając, Michał"'
Autor:
Howe, Nikolaus, Zajac, Michał, McKenzie, Ian, Hollinsworth, Oskar, Tseng, Tom, Bacon, Pierre-Luc, Gleave, Adam
Language model capabilities predictably improve from scaling a model's size and training data. Motivated by this, increasingly large language models have been trained, yielding an array of impressive capabilities. Yet these models are vulnerable to a
Externí odkaz:
http://arxiv.org/abs/2407.18213
Autor:
Wołczyk, Maciej, Cupiał, Bartłomiej, Ostaszewski, Mateusz, Bortkiewicz, Michał, Zając, Michał, Pascanu, Razvan, Kuciński, Łukasz, Miłoś, Piotr
Fine-tuning is a widespread technique that allows practitioners to transfer pre-trained capabilities, as recently showcased by the successful applications of foundation models. However, fine-tuning reinforcement learning (RL) models remains a challen
Externí odkaz:
http://arxiv.org/abs/2402.02868
Language model attacks typically assume one of two extreme threat models: full white-box access to model weights, or black-box access limited to a text generation API. However, real-world APIs are often more flexible than just text generation: these
Externí odkaz:
http://arxiv.org/abs/2312.14302
Class-incremental learning (CIL) is a particularly challenging variant of continual learning, where the goal is to learn to discriminate between all classes presented in an incremental fashion. Existing approaches often suffer from excessive forgetti
Externí odkaz:
http://arxiv.org/abs/2305.18806
Autor:
Zając, Michał, Deja, Kamil, Kuzina, Anna, Tomczak, Jakub M., Trzciński, Tomasz, Shkurti, Florian, Miłoś, Piotr
Diffusion models have achieved remarkable success in generating high-quality images thanks to their novel training procedures applied to unprecedented amounts of data. However, training a diffusion model from scratch is computationally expensive. Thi
Externí odkaz:
http://arxiv.org/abs/2303.15342
Autor:
Olko, Mateusz, Zając, Michał, Nowak, Aleksandra, Scherrer, Nino, Annadani, Yashas, Bauer, Stefan, Kuciński, Łukasz, Miłoś, Piotr
Inferring causal structure from data is a challenging task of fundamental importance in science. Observational data are often insufficient to identify a system's causal structure uniquely. While conducting interventions (i.e., experiments) can improv
Externí odkaz:
http://arxiv.org/abs/2211.13715
The ability of continual learning systems to transfer knowledge from previously seen tasks in order to maximize performance on new tasks is a significant challenge for the field, limiting the applicability of continual learning solutions to realistic
Externí odkaz:
http://arxiv.org/abs/2209.13900
Continual learning (CL) -- the ability to continuously learn, building on previously acquired knowledge -- is a natural requirement for long-lived autonomous reinforcement learning (RL) agents. While building such agents, one needs to balance opposin
Externí odkaz:
http://arxiv.org/abs/2105.10919
Autor:
Kurach, Karol, Raichuk, Anton, Stańczyk, Piotr, Zając, Michał, Bachem, Olivier, Espeholt, Lasse, Riquelme, Carlos, Vincent, Damien, Michalski, Marcin, Bousquet, Olivier, Gelly, Sylvain
Recent progress in the field of reinforcement learning has been accelerated by virtual learning environments such as video games, where novel algorithms and ideas can be quickly tested in a safe and reproducible manner. We introduce the Google Resear
Externí odkaz:
http://arxiv.org/abs/1907.11180
Recent work has shown that using unlabeled data in semi-supervised learning is not always beneficial and can even hurt generalization, especially when there is a class mismatch between the unlabeled and labeled examples. We investigate this phenomeno
Externí odkaz:
http://arxiv.org/abs/1904.03515