Filters








46,168 Hits in 2.3 sec

"... Von Der Alten Milchlingischen Burg Vnnd Stammhauß Mehr Nit Dann Die Rudera ..."

Susanne Gerschlauer, Michael Gottwald, Volker Hess, Christoph Röder
2013 Zenodo  
Volker Hess, Christoph Röder Ursprünglich als eine Weiterbildungsmaßnahme für ehrenamtliche Mitarbeiter der hessenarchäo logie im Landkreis Gießen geplant, hat sich die Untersuchung der vermeintlichen  ...  Lumda, Landkreis Gießen ________________________________________________________ "… von der alten Milchlingischen Burg vnnd Stammhauß mehr nit dann die rudera …" Susanne Gerschlauer, Michael Gottwald,  ... 
doi:10.5281/zenodo.1145800 fatcat:jbnhk34vsbbz7mx52apterjskq

The Human Kernel [article]

Andrew Gordon Wilson, Christoph Dann, Christopher G. Lucas, Eric P. Xing
2015 arXiv   pre-print
Bayesian nonparametric models, such as Gaussian processes, provide a compelling framework for automatic statistical modelling: these models have a high degree of flexibility, and automatically calibrated complexity. However, automating human expertise remains elusive; for example, Gaussian processes with standard kernels struggle on function extrapolation problems that are trivial for human learners. In this paper, we create function extrapolation problems and acquire human responses, and then
more » ... esign a kernel learning framework to reverse engineer the inductive biases of human learners across a set of behavioral experiments. We use the learned kernels to gain psychological insights and to extrapolate in human-like ways that go beyond traditional stationary and polynomial kernels. Finally, we investigate Occam's razor in human and Gaussian process based function learning.
arXiv:1510.07389v3 fatcat:gkklb66dl5dztd4fd5mhch4kve

Thoughts on Massively Scalable Gaussian Processes [article]

Andrew Gordon Wilson, Christoph Dann, Hannes Nickisch
2015 arXiv   pre-print
We introduce a framework and early results for massively scalable Gaussian processes (MSGP), significantly extending the KISS-GP approach of Wilson and Nickisch (2015). The MSGP framework enables the use of Gaussian processes (GPs) on billions of datapoints, without requiring distributed inference, or severe assumptions. In particular, MSGP reduces the standard O(n^3) complexity of GP learning and inference to O(n), and the standard O(n^2) complexity per test point prediction to O(1). MSGP
more » ... ves 1) decomposing covariance matrices as Kronecker products of Toeplitz matrices approximated by circulant matrices. This multi-level circulant approximation allows one to unify the orthogonal computational benefits of fast Kronecker and Toeplitz approaches, and is significantly faster than either approach in isolation; 2) local kernel interpolation and inducing points to allow for arbitrarily located data inputs, and O(1) test time predictions; 3) exploiting block-Toeplitz Toeplitz-block structure (BTTB), which enables fast inference and learning when multidimensional Kronecker structure is not present; and 4) projections of the input space to flexibly model correlated inputs and high dimensional data. The ability to handle many (m ≈ n) inducing points allows for near-exact accuracy and large scale kernel learning.
arXiv:1511.01870v1 fatcat:mjk7vb3ylfgbthwdhagrpuspya

Decoupling Learning Rules from Representations [article]

Philip S. Thomas and Christoph Dann and Emma Brunskill
2017 arXiv   pre-print
In the artificial intelligence field, learning often corresponds to changing the parameters of a parameterized function. A learning rule is an algorithm or mathematical expression that specifies precisely how the parameters should be changed. When creating an artificial intelligence system, we must make two decisions: what representation should be used (i.e., what parameterized function should be used) and what learning rule should be used to search through the resulting set of representable
more » ... ctions. Using most learning rules, these two decisions are coupled in a subtle (and often unintentional) way. That is, using the same learning rule with two different representations that can represent the same sets of functions can result in two different outcomes. After arguing that this coupling is undesirable, particularly when using artificial neural networks, we present a method for partially decoupling these two decisions for a broad class of learning rules that span unsupervised learning, reinforcement learning, and supervised learning.
arXiv:1706.03100v1 fatcat:7lkegq3qqje7hoeihg3hsp62ia

Reinforcement Learning with Feedback Graphs [article]

Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
2020 arXiv   pre-print
Next, using Lemma 11 from Dann et al.  ...  An application of Lemma 17 in Dann et al.  ... 
arXiv:2005.03789v1 fatcat:iyfcslvyqfdb5bguyb62wu54nm

Neural Active Learning with Performance Guarantees [article]

Pranjal Awasthi, Christoph Dann, Claudio Gentile, Ayush Sekhari, Zhilei Wang
2021 arXiv   pre-print
We investigate the problem of active learning in the streaming setting in non-parametric regimes, where the labels are stochastically generated from a class of functions on which we make no assumptions whatsoever. We rely on recently proposed Neural Tangent Kernel (NTK) approximation tools to construct a suitable neural embedding that determines the feature space the algorithm operates on and the learned model computed atop. Since the shape of the label requesting threshold is tightly related
more » ... the complexity of the function to be learned, which is a-priori unknown, we also derive a version of the algorithm which is agnostic to any prior knowledge. This algorithm relies on a regret balancing scheme to solve the resulting online model selection problem, and is computationally efficient. We prove joint guarantees on the cumulative regret and number of requested labels which depend on the complexity of the labeling function at hand. In the linear case, these guarantees recover known minimax results of the generalization error as a function of the label complexity in a standard statistical learning setting.
arXiv:2106.03243v1 fatcat:otgrek7fbbap3nglo2cipcy6gq

G+J Brand Sculpture: Visualisierung komplexer Markenbeziehungen

Christoph Danne
2014 MedienWirtschaft  
Daraus lassen sich dann auch weiterreichende markenspezifische Erkenntnisse ableiten -beispielsweise eine Art Frühwarnsystem zur Kundenbeziehung: Während große Nähe zum Konsumenten etwas Positives ist,  ... 
doi:10.15358/1613-0669-2014-1-24 fatcat:u66x6sc7o5fmrjpr2vtcsdnxfu

Memory Lens: How Much Memory Does an Agent Use? [article]

Christoph Dann, Katja Hofmann, Sebastian Nowozin
2016 arXiv   pre-print
We propose a new method to study the internal memory used by reinforcement learning policies. We estimate the amount of relevant past information by estimating mutual information between behavior histories and the current action of an agent. We perform this estimation in the passive setting, that is, we do not intervene but merely observe the natural behavior of the agent. Moreover, we provide a theoretical justification for our approach by showing that it yields an implementation-independent
more » ... wer bound on the minimal memory capacity of any agent that implement the observed policy. We demonstrate our approach by estimating the use of memory of DQN policies on concatenated Atari frames, demonstrating sharply different use of memory across 49 games. The study of memory as information that flows from the past to the current action opens avenues to understand and improve successful reinforcement learning algorithms.
arXiv:1611.06928v1 fatcat:zefqfsf665do7kzqfwnfwq74c4

Same Cause; Different Effects in the Brain [article]

Mariya Toneva, Jennifer Williams, Anand Bollu, Christoph Dann, Leila Wehbe
2022 arXiv   pre-print
To study information processing in the brain, neuroscientists manipulate experimental stimuli while recording participant brain activity. They can then use encoding models to find out which brain "zone" (e.g. which region of interest, volume pixel or electrophysiology sensor) is predicted from the stimulus properties. Given the assumptions underlying this setup, when stimulus properties are predictive of the activity in a zone, these properties are understood to cause activity in that zone. In
more » ... ecent years, researchers have used neural networks to construct representations that capture the diverse properties of complex stimuli, such as natural language or natural images. Encoding models built using these high-dimensional representations are often able to significantly predict the activity in large swathes of cortex, suggesting that the activity in all these brain zones is caused by stimulus properties captured in the representation. It is then natural to ask: "Is the activity in these different brain zones caused by the stimulus properties in the same way?" In neuroscientific terms, this corresponds to asking if these different zones process the stimulus properties in the same way. Here, we propose a new framework that enables researchers to ask if the properties of a stimulus affect two brain zones in the same way. We use simulated data and two real fMRI datasets with complex naturalistic stimuli to show that our framework enables us to make such inferences. Our inferences are strikingly consistent between the two datasets, indicating that the proposed framework is a promising new tool for neuroscientists to understand how information is processed in the brain.
arXiv:2202.10376v1 fatcat:yk7ixv32enfh5cer4cm65ebuhe

A Model Selection Approach for Corruption Robust Reinforcement Learning [article]

Chen-Yu Wei, Christoph Dann, Julian Zimmert
2021 arXiv   pre-print
Ashok Cutkosky, Christoph Dann, Abhimanyu Das, Claudio Gentile, Aldo Pacchiano, and Manish Purohit. Dynamic balancing for model selection in bandits and rl.  ...  Christoph Dann and Emma Brunskill. Sample complexity of episodic fixed-horizon reinforcement learning. In Conference on Neural Information Processing Systems, 2015.  ...  This can also be proved formally through the use of feedback graphs (Dann et al., 2020) .  ... 
arXiv:2110.03580v1 fatcat:us2atl3cybhbtekysz4uiqjmau

Agnostic Reinforcement Learning with Low-Rank MDPs and Rich Observations [article]

Christoph Dann, Yishay Mansour, Mehryar Mohri, Ayush Sekhari, Karthik Sridharan
2021 arXiv   pre-print
., 2017; Dann et al., 2019) and do not scale to the rich observation setting.  ...  ., 2017; Dann et al., 2018; Du et al., 2019a; Misra et al., 2020; Agarwal et al., 2020b) .  ... 
arXiv:2106.11519v1 fatcat:mh2pfxy62rfhfjf2z3prbutsje

Bayesian Time-of-Flight for Realtime Shape, Illumination and Albedo [article]

Amit Adam, Christoph Dann, Omer Yair, Shai Mazor, Sebastian Nowozin
2015 arXiv   pre-print
Parts of this work have been done when Christoph Dann interned at Microsoft Research Cambridge, UK.  ... 
arXiv:1507.06173v1 fatcat:nqx6medrhzfmvbcdmb2y7ygm4u

Sample Efficient Policy Search for Optimal Stopping Domains [article]

Karan Goel, Christoph Dann, Emma Brunskill
2017 arXiv   pre-print
Optimal stopping problems consider the question of deciding when to stop an observation-generating process in order to maximize a return. We examine the problem of simultaneously learning and planning in such domains, when data is collected directly from the environment. We propose GFSE, a simple and flexible model-free policy search method that reuses data for sample efficiency by leveraging problem structure. We bound the sample complexity of our approach to guarantee uniform convergence of
more » ... licy value estimates, tightening existing PAC bounds to achieve logarithmic dependence on horizon length for our setting. We also examine the benefit of our method against prevalent model-based and model-free approaches on 3 domains taken from diverse fields.
arXiv:1702.06238v2 fatcat:mhra62hjtzg4ze4wx6tczyihbu

Regret Bound Balancing and Elimination for Model Selection in Bandits and RL [article]

Aldo Pacchiano, Christoph Dann, Claudio Gentile, Peter Bartlett
2020 arXiv   pre-print
We propose a simple model selection approach for algorithms in stochastic bandit and reinforcement learning problems. As opposed to prior work that (implicitly) assumes knowledge of the optimal regret, we only require that each base algorithm comes with a candidate regret bound that may or may not hold during all rounds. In each round, our approach plays a base algorithm to keep the candidate regret bounds of all remaining base algorithms balanced, and eliminates algorithms that violate their
more » ... ndidate bound. We prove that the total regret of this approach is bounded by the best valid candidate regret bound times a multiplicative factor. This factor is reasonably small in several applications, including linear bandits and MDPs with nested function classes, linear bandits with unknown misspecification, and LinUCB applied to linear bandits with different confidence parameters. We further show that, under a suitable gap-assumption, this factor only scales with the number of base algorithms and not their complexity when the number of rounds is large enough. Finally, unlike recent efforts in model selection for linear stochastic bandits, our approach is versatile enough to also cover cases where the context information is generated by an adversarial environment, rather than a stochastic one.
arXiv:2012.13045v1 fatcat:qjjf57epszdkrazhupxyrbilqu

Policy Certificates: Towards Accountable Reinforcement Learning [article]

Christoph Dann, Lihong Li, Wei Wei, Emma Brunskill
2019 arXiv   pre-print
Correspondence to: Christoph Dann <cdann@cdann.net>. algorithms that can quantify and reveal their performance online during learning.  ...  The same is true for the stronger Uniform-PAC bounds (Dann et al., 2017) which hold for all jointly. • Supervised-learning style PAC bounds (Kearns & Singh, 2002; Jiang et al., 2017; Dann et al., 2018  ... 
arXiv:1811.03056v3 fatcat:l7qefuzasvhstibggnwudsgnnq
« Previous Showing results 1 — 15 out of 46,168 results