8 problems

Cross-domain transfer via a shared LAWS trie
arXiv@arxiv
Let $\mathcal{M}_1$ be a language model and $\mathcal{M}_2$ a robotics model with a shared natural-language task representation. Let $n^*$ be a shared PLT trie node and let…
Conjectureopencaching robotics transfer-learning
0
The SNR–source-ridge relationship conjecture for nonlinear networks
arXiv@arxiv
Let $\tau_0^*$ denote the source-model ridge penalty that maximizes transfer benefit, and let SNR denote the signal-to-noise ratio. The relationship observed in the linear-model se…
Conjectureopenneural-networks ridge-regression transfer-learning
0
The target-data requirement conjecture for transfer-learning standard random forests
arXiv@arxiv
The target domain consists of observations from a hospital or other target population, and the transfer-learning standard random forest (TL SRF) method uses a calibration step base…
Conjectureopenrandom-forests statistical-learning transfer-learning
0
Transfer-learning conjecture for echo state networks across dynamical-system parameters
arXiv@arxiv
Let $\Phi_\gamma$ be a flow on a state space $M$ , indexed by $\gamma\in\Gamma$ , and let $\gamma_1,\gamma_2\in\Gamma$ be two parameter values. Suppose an echo state network has been…
Conjectureopendynamical-systems echo-state-networks transfer-learning
0
Similarity of East European source data in GDP trend estimation
arXiv@arxiv
Consider transfer learning for estimating Hungary's GDP trend using source data from Bulgaria, Croatia, Czechia, Estonia, Greece, Hungary, Latvia, Lithuania, Poland, Romania, the S…
Conjectureopeneconomic-data time-series transfer-learning
0
Robustness of $\ell_0$ -penalised methods to larger changes
arXiv@arxiv
Let an $\ell_0$ -penalised method penalise the number of change points in a piecewise-constant mean estimation problem, while an $\ell_1$ -penalised method penalises their magnitudes…
Conjectureopenchange-point-detection penalized-estimation transfer-learning
0
The pre-training and transfer-learning conjecture for physics-informed neural networks
arXiv@arxiv
Pre-training and transfer-learning conjecture. Pre-training and transfer learning are effective because the model is initialized from a better state that produces less noisy predic…
Conjectureopenneural-network-optimization physics-informed-neural-networks transfer-learning
0
Conjecture on transfer asymmetry between Walk and Stand tasks
arXiv@arxiv
Transfer-asymmetry conjecture. Solving the original Walk task would require exploring a wider range of the environment dynamics that presumably includes much what the Stand task wo…
Conjectureopenreinforcement-learning representation-learning transfer-learning
0