Massachusetts Institute of Technology

About: Massachusetts Institute of Technology is a education organization based out in Cambridge, Massachusetts, United States. It is known for research contribution in the topics: Population & Laser. The organization has 116795 authors who have published 268000 publications receiving 18272025 citations. The organization is also known as: MIT & M.I.T..

Networks for approximation and learning

Abstract: The problem of the approximation of nonlinear mapping, (especially continuous mappings) is considered. Regularization theory and a theoretical framework for approximation (based on regularization techniques) that leads to a class of three-layer networks called regularization networks are discussed. Regularization networks are mathematically related to the radial basis functions, mainly used for strict interpolation tasks. Learning as approximation and learning as hypersurface reconstruction are discussed. Two extensions of the regularization approach are presented, along with the approach's corrections to splines, regularization, Bayes formulation, and clustering. The theory of regularization networks is generalized to a formulation that includes task-dependent clustering and dimensionality reduction. Applications of regularization networks are discussed. >

Alignment by Maximization of Mutual Information

Abstract: A new information-theoretic approach is presented for finding the pose of an object in an image. The technique does not require information about the surface properties of the object, besides its shape, and is robust with respect to variations of illumination. In our derivation few assumptions are made about the nature of the imaging process. As a result the algorithms are quite general and may foreseeably be used in a wide variety of imaging situations. Experiments are presented that demonstrate the approach registering magnetic resonance (MR) images, aligning a complex 3D object model to real scenes including clutter and occlusion, tracking a human head in a video sequence and aligning a view-based 2D object model to real images. The method is based on a formulation of the mutual information between the model and the image. As applied here the technique is intensity-based, rather than feature-based. It works well in domains where edge or gradient-magnitude based methods have difficulty, yet it is more robust than traditional correlation. Additionally, it has an efficient implementation that is based on stochastic approximation.

Towards Deep Learning Models Resistant to Adversarial Attacks.

Abstract: Recent work has demonstrated that deep neural networks are vulnerable to adversarial examples—inputs that are almost indistinguishable from natural data and yet classified incorrectly by the network. In fact, some of the latest findings suggest that the existence of adversarial attacks may be an inherent weakness of deep learning models. To address this problem, we study the adversarial robustness of neural networks through the lens of robust optimization. This approach provides us with a broad and unifying view on much of the prior work on this topic. Its principled nature also enables us to identify methods for both training and attacking neural networks that are reliable and, in a certain sense, universal. In particular, they specify a concrete security guarantee that would protect against any adversary. These methods let us train networks with significantly improved resistance to a wide range of adversarial attacks. They also suggest the notion of security against a first-order adversary as a natural and broad security guarantee. We believe that robustness against such well-defined classes of adversaries is an important stepping stone towards fully resistant deep learning models. Code and pre-trained models are available at this https URL and this https URL.

How to play ANY mental game

Abstract: We present a polynomial-time algorithm that, given as a input the description of a game with incomplete information and any number of players, produces a protocol for playing the game that leaks no partial information, provided the majority of the players is honest. Our algorithm automatically solves all the multi-party protocol problems addressed in complexity-based cryptography during the last 10 years. It actually is a completeness theorem for the class of distributed protocols with honest majority. Such completeness theorem is optimal in the sense that, if the majority of the players is not honest, some protocol problems have no efficient solution [C].

The impact of microRNAs on protein output

Abstract: MicroRNAs are endogenous ∼23-nucleotide RNAs that can pair to sites in the messenger RNAs of protein-coding genes to downregulate the expression from these messages. MicroRNAs are known to influence the evolution and stability of many mRNAs, but their global impact on protein output had not been examined. Here we use quantitative mass spectrometry to measure the response of thousands of proteins after introducing microRNAs into cultured cells and after deleting mir-223 in mouse neutrophils. The identities of the responsive proteins indicate that targeting is primarily through seed-matched sites located within favourable predicted contexts in 3′ untranslated regions. Hundreds of genes were directly repressed, albeit each to a modest degree, by individual microRNAs. Although some targets were repressed without detectable changes in mRNA levels, those translationally repressed by more than a third also displayed detectable mRNA destabilization, and, for the more highly repressed targets, mRNA destabilization usually comprised the major component of repression. The impact of microRNAs on the proteome indicated that for most interactions microRNAs act as rheostats to make fine-scale adjustments to protein output. MicroRNAs can regulate gene expression by either inhibiting translation of a messenger RNA, or inducing its degradation. While previous studies have measured regulation at the mRNA level, it was unknown how much regulation occurred at the protein level. Now two groups led by David Bartel and Nikolaus Rajewsky have used variants of the technique known as SILAC (stable isotope labelling with amino acids in cell culture) to measure proteome-wide changes in protein level as a function of expression of endogenous and exogenous microRNAs. They find that while microRNAs can directly repress the translation of hundreds of genes, additional indirect effects result in changes in expression of thousands of genes. Many of the changes observed are less than twofold in magnitude, however, indicating either directly or indirectly, microRNAs can act as rheostats to fine-tune protein synthesis to match the needs of the cell at any given time. In one of two studies, a technique known as SILAC is used to measure, on a large scale, changes in protein level as a function of expression of endogenous and exogenous miRNAs. It is found that although miRNAs directly repress the translation of hundreds of genes, additional indirect effects result in changes in expression of thousands of genes.