Zobrazeno 1 - 10
of 331
pro vyhledávání: '"Shafto P"'
When decisions are made at high frequency, traditional reinforcement learning (RL) methods struggle to accurately estimate action values. In turn, their performance is inconsistent and often poor. Whether the performance of distributional RL (DRL) ag
Externí odkaz:
http://arxiv.org/abs/2410.11022
Autor:
Li, Zhaobin, Shafto, Patrick
Intent obfuscation is a common tactic in adversarial situations, enabling the attacker to both manipulate the target system and avoid culpability. Surprisingly, it has rarely been implemented in adversarial attacks on machine learning systems. We are
Externí odkaz:
http://arxiv.org/abs/2408.02674
Tabular data is common yet typically incomplete, small in volume, and access-restricted due to privacy concerns. Synthetic data generation offers potential solutions. Many metrics exist for evaluating the quality of synthetic tabular data; however, w
Externí odkaz:
http://arxiv.org/abs/2403.10424
Autor:
Hao, Xiaoran, Shafto, Patrick
Variational auto-encoders are powerful probabilistic models in generative tasks but suffer from generating low-quality samples which are caused by the holes in the prior. We propose the Coupled Variational Auto-Encoder (C-VAE), which formulates the V
Externí odkaz:
http://arxiv.org/abs/2306.02565
Evolution of beliefs of a society are a product of interactions between people (horizontal transmission) in the society over generations (vertical transmission). Researchers have studied both horizontal and vertical transmission separately. Extending
Externí odkaz:
http://arxiv.org/abs/2205.13587
The goal of explainable Artificial Intelligence (XAI) is to generate human-interpretable explanations, but there are no computationally precise theories of how humans interpret AI generated explanations. The lack of theory means that validation of XA
Externí odkaz:
http://arxiv.org/abs/2205.08452
Autor:
Lu, Chi-Ken, Shafto, Patrick
Deep Gaussian Process (DGP) as a model prior in Bayesian learning intuitively exploits the expressive power in function composition. DGPs also offer diverse modeling capabilities, but inference is challenging because marginalization in latent functio
Externí odkaz:
http://arxiv.org/abs/2203.07411
Autor:
Aad, G., Akimov, A. V., Khoury, K. Al, Aleksa, M., Andeen, T., Anelli, C., Aranzabal, N., Armijo, C., Bagulia, A., Ban, J., Barillari, T., Bellachia, F., Benoit, M., Bernon, F., Berthold, A., Bervas, H., Besin, D., Betti, A., Bianga, Y., Biaut, M., Boline, D., Boudreau, J., Bouedo, T., Braam, N., Bret, M. Cano, Brooijmans, G., Cai, H., Camincher, C., Camplani, A., Cap, S., Carbone, A., Carter, J. W. S., Chekulaev, S. V., Chen, H., Chen, K., Chevillot, N., Citterio, M., Cleland, B., Constable, M., de Jong, S., Deiana, A. M., Delmastro, M., Deng, B., Deschamps, H., Diaconu, C., Dik, A., Dinkespiler, B., Dayot, N. Dumont, Emerman, A., Enari, Y., Falke, P. J., Farrell, J., Fielitz, W., Fortin, E., Fragnaud, J., Franchino, S., Gantel, L., Gigliotti, K., Gong, D., Grabas, A., Grohs, P., Guettouche, N., Guillemin, T., Guo, D., Guo, J., Hasley, L., Hayes, C., Hentges, R., Hervas, L., Hils, M., Hobbs, J., Hoffman, A., Hoffmann, D., Horn, P., Hryn'ova, T., Iconomidou-Fayard, L., Iguchi, R., James, T., Johns, K., Junkermann, T., Kahra, C., Kay, E. F., Keeler, R., Haghighat, S. Ketabchi, Kinget, P., Knoops, E., Kolbasin, A., Krieger, P., Kuppambatti, J., Kurchaninov, L. L., Ladygin, E., Lafrasse, S., Landon, M. P. J., Lanni, F., Latorre, S., Laugier, D., Lazzaroni, M., Le, X., Bourlout, P. Le, Lee, C. A., Lefebvre, M., Leite, M. A. L., Leroy, C., Li, X., Li, Z., Liang, F., Liu, H., Liu, C., Liu, T., Ma, H., Ma, L. L., Mahon, D. J., Mallik, U., Mansoulie, B., Maslennikov, A. L., Matsuzawa, N., McPherson, R. A., Menke, S., Milic, A., Minami, Y., Molina, E., Monnier, E., Morange, N., Morvaj, L., Mueller, J., Mwewa, C., Narayan, R., Nikiforou, N., Ochoa, I., Oishi, R., Damazio, D. Oliveira, Owen, R. E., Pancake, C., Panchal, D. K., Perrot, G., Pleier, M. -A., Poffenberger, P., Porter, R., Quan, S., Rabel, J., Roy, A., Rutherfoord, J. P., Sabatini, F., Salomon, F., Sauvan, E., Schaffer, A. C., Schamberger, R. D., Schwemling, Ph., Secord, C., Selem, L., Sexton, K., Shafto, E., Oliveira, M. V. Silva, Simion, S., Singh, S., Sippach, W., Snesarev, A. A., Snyder, S., Spalla, M., Stärz, S., Straessner, A., Strizenec, P., Stroynowski, R., Sulin, V. V., Tanaka, J., Tang, S., Tapprogge, S., Tartarelli, G. F., Tateno, G., Terashi, K., Tisserant, S., Tompkins, D., Unal, G., Unal, M., Uno, K., Vallier, A., de Souza, S. Vieira, Walker, R., Wang, Q., Wang, C., Wang, R., Wessels, M., Wingerter-Seez, I., Wolniewicz, K., Wu, W., Xiandong, Z., Xu, R., Xu, H., Yamamoto, S., Yang, Y., Ye, J., Zaghia, H., Zang, J., Zhang, T., Zhu, H. L., Zhulanov, V., Zonca, E., Zuk, G.
Publikováno v:
2022 JINST 17 P05024
The Phase-I trigger readout electronics upgrade of the ATLAS Liquid Argon calorimeters enhances the physics reach of the experiment during the upcoming operation at increasing Large Hadron Collider luminosities. The new system, installed during the s
Externí odkaz:
http://arxiv.org/abs/2202.07384
Optimal transport (OT) formalizes the problem of finding an optimal coupling between probability measures given a cost matrix. The inverse problem of inferring the cost given a coupling is Inverse Optimal Transport (IOT). IOT is less well understood
Externí odkaz:
http://arxiv.org/abs/2112.09754
Autor:
Lu, Chi-Ken, Shafto, Patrick
Publikováno v:
Entropy 2021, 23(11), 1387
It is desirable to combine the expressive power of deep learning with Gaussian Process (GP) in one expressive Bayesian learning model. Deep kernel learning showed success in adopting a deep network for feature extraction followed by a GP used as func
Externí odkaz:
http://arxiv.org/abs/2110.00568