Constrained Image Generation Using Binarized Neural Networks with Decision Procedures

@inproceedings{Korneev2018ConstrainedIG,
  title={Constrained Image Generation Using Binarized Neural Networks with Decision Procedures},
  author={Svyatoslav Korneev and Nina Narodytska and Luca Pulina and Armando Tacchella and Nikolaj S. Bj{\o}rner and Shmuel Sagiv},
  booktitle={International Conference on Theory and Applications of Satisfiability Testing},
  year={2018},
  url={https://api.semanticscholar.org/CorpusID:3545062}
}
This work considers the problem of binary image generation with given properties and uses a binarized neural network to approximate a PDE solver to solve the problem, showing that this problem can be tackled using decision procedures.
[PDF] Semantic Reader
13 Citations
Background Citations
3
Methods Citations
5

13 Citations

Mechanical Engineering 6-4-2019 Encoding Invariances in Deep Generative Models

A new generative modeling approach, InvNet, that can efficiently model data spaces with known invariances and devise an adversarial training algorithm to encode them into data distribution is proposed.

Physics-constrained deterministic solution of time-dependent partial differential equations using deep convolutional encoder-decoders

This thesis deals with exploring the application of a specific branch of AI, or a specific tool, Deep Learning to solving real world engineering problems governed by Partial Differential equations which otherwise had been difficult to solve using existing methods till date.

Precise Quantitative Analysis of Binarized Neural Networks: A BDD-based Approach

A novel algorithmic approach for encoding BNNs as Binary Decision Diagrams (BDDs), a widely studied model in formal verification and knowledge representation, is proposed, which translates the input-output relation of blocks in Bnns to cardinality constraints, which are then encoded by BDDs.

Formal Analysis of Deep Binarized Neural Networks

This work designs a framework for analysis of properties of neural networks that can be represented and analyzed using well-developed means of Boolean Satisfiability and Integer Linear Programming and presents an exact representation of a binarized neural network as a Boolean formula.

Encoding Invariances in Deep Generative Models

A new generative modeling approach, InvNet, that can efficiently model data spaces with known invariances and devise an adversarial training algorithm to encode them into data distribution is proposed.

BSTC: A Novel Binarized-Soft-Tensor-Core Design for Accelerating Bit-Based Approximated Neural Nets

Experiments show that the Singular-Binarized-Neural-Network (SBNN) design can achieve over 1000x speedup for raw inference latency over the state-of-the-art full-precision BNN inference for AlexNet on GPUs.

Accelerating Binarized Neural Networks via Bit-Tensor-Cores in Turing GPUs

It is shown that the stride of memory access can significantly affect performance delivery and a data-format co-design is highly desired to support the Tensorcore-accelerated BNN design for achieving superior performance than existing software solutions without tensorcores.

BDD4BNN: A BDD-based Quantitative Analysis Framework for Binarized Neural Networks

A quantitative framework for Binarized Neural Networks, the 1-bit quantization of general real-numbered neural networks, is developed where precise and comprehensive analysis of BNNs can be performed and is demonstrated by providing quantitative robustness analysis and interpretability.

ESampler: Boosting sampling of satisfying assignments for Boolean formulas via derivation

ESampler: Efficient Sampling of Satisfying Assignments for Boolean Formulas

This work proposes a novel approach to derive a large set of satisfying assignments from a given one in an efficient way and could be integrated into the existing SAT sam-plers.

26 References

GANs for Biological Image Synthesis

By interpolating across the latent space, GANs can mimic the known changes in protein localization that occur through time during the cell cycle, allowing us to predict temporal evolution from static images.

Gated XNOR Networks: Deep Neural Networks with Ternary Weights and Activations under a Unified Discretization Framework

A multistep neuronal activation discretization method and a derivative approximation technique are presented that enable the implementing the back propagation algorithm on discrete DNNs and a discrete state transition (DST) methodology is proposed to constrain the weights in a discrete space without saving the hidden weights.

Verifying Properties of Binarized Deep Neural Networks

This paper proposes a rigorous way of verifying properties of a popular class of neural networks, Binarized Neural Networks, using the well-developed means of Boolean satisfiability, and creates a construction that creates a representation of a binarized neural network as a Boolean formula.

Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks

Results show that the novel, scalable, and efficient technique presented can successfully prove properties of networks that are an order of magnitude larger than the largest networks verified using existing methods.

Safety Verification of Deep Neural Networks

A novel automated verification framework for feed-forward multi-layer neural networks based on Satisfiability Modulo Theory (SMT) is developed, which defines safety for an individual decision in terms of invariance of the classification within a small neighbourhood of the original image.

Binarized Neural Networks

Near state-of-the-art results with BDNNs are obtained on the permutation-invariant MNIST, CIFAR-10 and SVHN datasets.

Freehand Ultrasound Image Simulation with Spatially-Conditioned Generative Adversarial Networks

This paper introduces a novel neural network architecture that can sample anatomically accurate images conditionally on spatial position of the (real or mock) freehand ultrasound probe, based on conditional generative adversarial networks (GANs), and presents what it believes are state-of-the-art visually realistic ultrasound images simulated by the proposed GAN architecture.

Verification of Binarized Neural Networks

The problem of formal verification of Binarized Neural Networks, which have recently been proposed as a power-efficient alternative to more traditional learning networks, is studied and it is proved that the problem of optimal factoring is NP-hard.

Enabling Dark Energy Science with Deep Generative Models of Galaxy Images

This work considers variations on conditional variational autoencoder and introduces a new adversarial objective for training of conditional generative networks that suggests a reliable alternative to the acquisition of expensive high quality observations for generating the calibration data needed by the next generation of cosmological surveys.

Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

This work introduces a class of CNNs called deep convolutional generative adversarial networks (DCGANs), that have certain architectural constraints, and demonstrates that they are a strong candidate for unsupervised learning.