Overcoming catastrophic forgetting in neural networks von Kirkpatrick, James, Pascanu, Razvan, Rabinowitz, Neil, Veness, Joel, Desjardins, Guillaume, Rusu, Andrei A., Milan, Kieran, Quan, John, Ramalho, Tiago, Grabska-Barwinska, Agnieszka, Hassabis, Demis, Clopath, Claudia, Kumaran, Dharshan, Hadsell, Raia

Volltext

Vector-based navigation using grid-like representations in artificial agents von Banino, Andrea, Barry, Caswell, Uria, Benigno, Blundell, Charles, Lillicrap, Timothy, Mirowski, Piotr, Pritzel, Alexander, Chadwick, Martin J., Degris, Thomas, Modayil, Joseph, Wayne, Greg, Soyer, Hubert, Viola, Fabio, Zhang, Brian, Goroshin, Ross, Rabinowitz, Neil, Pascanu, Razvan, Beattie, Charlie, Petersen, Stig, Sadik, Amir, Gaffney, Stephen, King, Helen, Kavukcuoglu, Koray, Hassabis, Demis, Hadsell, Raia, Kumaran, Dharshan

Volltext

Task switching can make AI lose the ability to learn von Lyle, Clare, Pascanu, Razvan

Volltext

Switching between tasks can cause AI to lose the ability to learn von Lyle, Clare, Pascanu, Razvan

Volltext

Embracing Change: Continual Learning in Deep Neural Networks von Hadsell, Raia, Rao, Dushyant, Rusu, Andrei A., Pascanu, Razvan

Volltext

A neurodynamical model for working memory von Pascanu, Razvan, Jaeger, Herbert

Volltext

Learned-Norm Pooling for Deep Feedforward and Recurrent Neural Networks von Gulcehre, Caglar, Cho, Kyunghyun, Pascanu, Razvan, Bengio, Yoshua

Volltext

Correlation Based Semantic Transfer with Application to Domain Adaptation von Calnegru, Florina Cristina, Shawe-Taylor, John, Kokkinos, Iasonas, Pascanu, Razvan

Volltext

REPLY TO HUSZÁR: The elastic weight consolidation penalty is empirically valid von Kirkpatrick, James, Pascanu, Razvan, Rabinowitz, Neil, Veness, Joel, Desjardins, Guillaume, Rusu, Andrei A., Milan, Kieran, Quan, John, Ramalho, Tiago, Grabska-Barwinska, Agnieszka, Hassabis, Demis, Clopath, Claudia, Kumaran, Dharshan, Hadsell, Raia

Volltext

Deep Grokking: Would Deep Neural Networks Generalize Better? von Fan, Simin, Pascanu, Razvan, Jaggi, Martin

Volltext bestellen

Latent Space Representations of Neural Algorithmic Reasoners von Mirjanić, Vladimir V, Pascanu, Razvan, Veličković, Petar

Volltext bestellen

Contextual tag inference von Mandel, Michael I., Pascanu, Razvan, Eck, Douglas, Bengio, Yoshua, Aiello, Luca M., Schifanella, Rossano, Menczer, Filippo

Volltext

Round and Round We Go! What makes Rotary Positional Encodings useful? von Barbero, Federico, Vitvitskyi, Alex, Perivolaropoulos, Christos, Pascanu, Razvan, Veličković, Petar

Volltext bestellen

softmax is not enough (for sharp out-of-distribution) von Veličković, Petar, Perivolaropoulos, Christos, Barbero, Federico, Pascanu, Razvan

Volltext bestellen

Investigating Low-Rank Training in Transformer Language Models: Efficiency and Scaling Analysis von Wei, Xiuying, Moalla, Skander, Pascanu, Razvan, Gulcehre, Caglar

Volltext bestellen

Building on Efficient Foundations: Effectively Training LLMs with Structured Feedforward Layers von Wei, Xiuying, Moalla, Skander, Pascanu, Razvan, Gulcehre, Caglar

Volltext bestellen

Attention as a Hypernetwork von Schug, Simon, Kobayashi, Seijin, Akram, Yassir, Sacramento, João, Pascanu, Razvan

Volltext bestellen

No Representation, No Trust: Connecting Representation, Collapse, and Trust Issues in PPO von Moalla, Skander, Miele, Andrea, Pyatko, Daniil, Pascanu, Razvan, Gulcehre, Caglar

Volltext bestellen

Asynchronous Algorithmic Alignment with Cocycles von Dudzik, Andrew, von Glehn, Tamara, Pascanu, Razvan, Veličković, Petar

Volltext bestellen

Learning to Modulate pre-trained Models in RL von Schmied, Thomas, Hofmarcher, Markus, Paischer, Fabian, Pascanu, Razvan, Hochreiter, Sepp

Volltext bestellen