Follow
Hidenori Tanaka
Hidenori Tanaka
Other names田中 秀宣
Group Leader, CBS-NTT Program in "Physics of Intelligence", Harvard University
Verified email at fas.harvard.edu - Homepage
Title
Cited by
Cited by
Year
Pruning neural networks without any data by iteratively conserving synaptic flow
H Tanaka, D Kunin, DLK Yamins, S Ganguli
NeurIPS (Advances in Neural Information Processing Systems), 2020
5832020
From deep learning to mechanistic understanding in neuroscience: the structure of retinal prediction
H Tanaka, A Nayebi, N Maheswaranathan, L McIntosh, SA Baccus, ...
NeurIPS (Advances in Neural Information Processing Systems), 2019
722019
Spatial gene drives and pushed genetic waves
H Tanaka, HA Stone, DR Nelson
PNAS (Proceedings of the National Academy of Sciences), 2017
722017
Neural Mechanics: Symmetry and Broken Conservation Laws in Deep Learning Dynamics
D Kunin, J Sagastuy-Brena, S Ganguli, DLK Yamins, H Tanaka
ICLR (International Conference on Learning and Representations), 2020
682020
Interpreting the retinal neural code for natural scenes: From computations to neurons
N Maheswaranathan*, LT McIntosh*, H Tanaka*, S Grant*, DB Kastner, ...
Neuron 111 (17), 2742-2755. e4, 2023
67*2023
Beyond BatchNorm: towards a unified understanding of normalization in deep learning
ES Lubana, R Dick, H Tanaka
NeurIPS (Advances in Neural Information Processing Systems) 34, 4778-4791, 2021
412021
Mechanistic mode connectivity
ES Lubana, EJ Bigelow, RP Dick, D Krueger, H Tanaka
International Conference on Machine Learning, 22965-23004, 2023
40*2023
Hot particles attract in a cold bath
H Tanaka, AA Lee, MP Brenner
Physical Review Fluids, 2017
312017
Noether’s learning dynamics: Role of symmetry breaking in neural networks
H Tanaka, D Kunin
NeurIPS (Advances in Neural Information Processing Systems) 34, 25646-25660, 2021
26*2021
Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
S Jain, R Kirk, ES Lubana, RP Dick, H Tanaka, E Grefenstette, ...
ICLR (International Conference on Learning and Representations), 2023
212023
Compositional abilities emerge multiplicatively: Exploring diffusion models on a synthetic task
M Okawa, ES Lubana, R Dick, H Tanaka
Advances in Neural Information Processing Systems 36, 2024
172024
Rethinking the limiting dynamics of SGD: modified loss, phase space oscillations, and anomalous diffusion
D Kunin, J Sagastuy-Brena, L Gillespie, E Margalit, H Tanaka, S Ganguli, ...
arXiv, 2021
17*2021
A lexical approach for identifying behavioural action sequences
G Reddy, L Desban, H Tanaka, J Roussel, O Mirat, C Wyart
PLoS computational biology 18 (1), e1009672, 2022
162022
Non-Hermitian quasilocalization and ring attractor neural networks
H Tanaka, DR Nelson
Physical Review E, 2019
162019
What shapes the loss landscape of self-supervised learning?
L Ziyin, ES Lubana, M Ueda, H Tanaka
ICLR (International Conference on Learning and Representations), 2022
152022
Mutation at Expanding Front of Self-Replicating Colloidal Clusters
H Tanaka, Z Zeravcic, MP Brenner
Physical Review Letters, 2016
112016
How capable can a transformer become? a study on synthetic, interpretable tasks
R Ramesh, ES Lubana, M Khona, RP Dick, H Tanaka
arXiv preprint arXiv:2311.12997, 2023
52023
Quenched metastable vortex states in Sr 2 RuO 4
D Shibata, H Tanaka, S Yonezawa, T Nojima, Y Maeno
Physical Review B, 2015
42015
Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
M Khona, M Okawa, J Hula, R Ramesh, ..., ES Lubana, H Tanaka
arXiv preprint arXiv:2402.07757, 2024
12024
CORNN: Convex optimization of recurrent neural networks for rapid inference of neural dynamics
F Dinc, A Shai, M Schnitzer, H Tanaka
Advances in Neural Information Processing Systems 36, 51273-51301, 2023
12023
The system can't perform the operation now. Try again later.
Articles 1–20