Zobrazeno 1 - 10
of 1 716
pro vyhledávání: '"Ziegler, A. M."'
Autor:
Hubinger, Evan, Denison, Carson, Mu, Jesse, Lambert, Mike, Tong, Meg, MacDiarmid, Monte, Lanham, Tamera, Ziegler, Daniel M., Maxwell, Tim, Cheng, Newton, Jermyn, Adam, Askell, Amanda, Radhakrishnan, Ansh, Anil, Cem, Duvenaud, David, Ganguli, Deep, Barez, Fazl, Clark, Jack, Ndousse, Kamal, Sachan, Kshitij, Sellitto, Michael, Sharma, Mrinank, DasSarma, Nova, Grosse, Roger, Kravec, Shauna, Bai, Yuntao, Witten, Zachary, Favaro, Marina, Brauner, Jan, Karnofsky, Holden, Christiano, Paul, Bowman, Samuel R., Graham, Logan, Kaplan, Jared, Mindermann, Sören, Greenblatt, Ryan, Shlegeris, Buck, Schiefer, Nicholas, Perez, Ethan
Humans are capable of strategically deceptive behavior: behaving helpfully in most situations, but then behaving very differently in order to pursue alternative objectives when given the opportunity. If an AI system learned such a deceptive strategy,
Externí odkaz:
http://arxiv.org/abs/2401.05566
Publikováno v:
Archiv der Mathematik 121 (2023), 559-601
We survey Jean-Louis Loday's vertex description of the associahedron, and its far reaching influence in combinatorics, discrete geometry and algebra. We present in particular four topics were it plays a central role: lattice congruences of the weak o
Externí odkaz:
http://arxiv.org/abs/2305.08471
Autor:
Sinn, Rainer, Ziegler, Günter M.
In his first mathematical paper, published in 1895 when he was 18, Edmund Landau suggested a new way to determine the winner of a chess tournament by not simply adding for each player the fixed number of points they would get for each win or draw, bu
Externí odkaz:
http://arxiv.org/abs/2210.17300
Autor:
Ziegler, Daniel M., Nix, Seraphina, Chan, Lawrence, Bauman, Tim, Schmidt-Nielsen, Peter, Lin, Tao, Scherlis, Adam, Nabeshima, Noa, Weinstein-Raun, Ben, de Haas, Daniel, Shlegeris, Buck, Thomas, Nate
In the future, powerful AI systems may be deployed in high-stakes settings, where a single failure could be catastrophic. One technique for improving AI safety in high-stakes settings is adversarial training, which uses an adversary to generate examp
Externí odkaz:
http://arxiv.org/abs/2205.01663
Autor:
Wu, Jeff, Ouyang, Long, Ziegler, Daniel M., Stiennon, Nisan, Lowe, Ryan, Leike, Jan, Christiano, Paul
A major challenge for scaling machine learning is training models to perform tasks that are very difficult or time-consuming for humans to evaluate. We present progress on this problem on the task of abstractive summarization of entire fiction novels
Externí odkaz:
http://arxiv.org/abs/2109.10862
Autor:
Benavides, Erika, Krecioch, Joseph R., Connolly, Roger T., Allareddy, Trishul, Buchanan, Allison, Spelic, David, O’Brien, Kelly K., Keels, Martha Ann, Mascarenhas, Ana Karina, Duong, Mai-Ly, Aerne-Bowe, Mickie J., Ziegler, Kathleen M., Lipman, Ruth D.
Publikováno v:
In The Journal of the American Dental Association April 2024 155(4):280-293
Autor:
Lane, Bennett H., Rea, David J., Gottula, Adam L., Cathers, Andrew D., Ziegler, Ryan M., Latimer, Andrew J., Danielson, Kyle R., Theiling, B. Jason, Froehle, Craig M., Hinckley, William R.
Publikováno v:
In Air Medical Journal March-April 2024 43(2):111-115
Autor:
Henighan, Tom, Kaplan, Jared, Katz, Mor, Chen, Mark, Hesse, Christopher, Jackson, Jacob, Jun, Heewoo, Brown, Tom B., Dhariwal, Prafulla, Gray, Scott, Hallacy, Chris, Mann, Benjamin, Radford, Alec, Ramesh, Aditya, Ryder, Nick, Ziegler, Daniel M., Schulman, John, Amodei, Dario, McCandlish, Sam
We identify empirical scaling laws for the cross-entropy loss in four domains: generative image modeling, video modeling, multimodal image$\leftrightarrow$text models, and mathematical problem solving. In all cases autoregressive Transformers smoothl
Externí odkaz:
http://arxiv.org/abs/2010.14701
Autor:
Stiennon, Nisan, Ouyang, Long, Wu, Jeff, Ziegler, Daniel M., Lowe, Ryan, Voss, Chelsea, Radford, Alec, Amodei, Dario, Christiano, Paul
As language models become more powerful, training and evaluation are increasingly bottlenecked by the data and metrics used for a particular task. For example, summarization models are often trained to predict human reference summaries and evaluated
Externí odkaz:
http://arxiv.org/abs/2009.01325
Robertson (1988) suggested a model for the realization space of a convex d-dimensional polytope and an approach via the implicit function theorem, which -- in the case of a full rank Jacobian -- proves that the realization space is a manifold of dime
Externí odkaz:
http://arxiv.org/abs/2007.00645