Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,291 papers · 148 categories

Trend · papers per month

199398597796 · Jun 202019922001200920182026
48 results for Neural Programmer

Deep neural networks have achieved impressive supervised classification performance in many tasks including image recognition, speech recognition, and sequence to sequence learning. However, this success has not been translated to applications like question answering that may involve complex arithmetic and logic reason…

2015-11-16abs ↗pdf ↗

FixyNN improves energy efficiency of mobile computer vision tasks.

problem High energy consumption of state-of-the-art CNN models on mobile devices.
method Fixed-weight feature extractor and programmable CNN accelerator for transfer learning.
result Achieved up to 26.6 TOPS/W energy efficiency, nearly 2x more efficient than conventional accelerators.

Paper reduces AI complexity with pre-defined sparsity and hardware acceleration.

problem Reduction of computational and storage complexity in neural networks.
method Pre-defined sparsity and hardware acceleration architecture.
result Significant reduction in storage and computational complexity (5X+ reduction) without significant performance loss.

Study evaluates training programs for unemployed in Belgium using machine learning.

problem Determining which training programs are most effective for unemployed individuals in Belgium.
method Used Modified Causal Forests, a causal machine learning estimator, to analyze data from unemployed in Belgium.
result There is significant heterogeneity in the effectiveness of different training programs for unemployed individuals in Belgium.

Researchers show NN-based communication algorithms can be implemented on hardware without significant performance loss.

problem Reducing complexity and improving performance of NN-based communication algorithms for practical hardware implementation.
method Implementation of NN-based algorithms in fixed-point arithmetic with quantized weights on specialized hardware (FPGAs, ASICs).
result It is possible to implement NN-based algorithms in fixed-point arithmetic with quantized weights on hardware without significant performance loss.

Graph neural networks improve charged particle tracking on FPGAs.

problem Charged particle trajectory determination in high interaction density conditions.
method Graph neural networks (GNNs) embedded in tracker data as graphs, classifying edges as track segments.
result GNNs implemented on FPGAs for charged particle tracking, enabling future HL-LHC experiments.

A fast method for learning MZI parameters in optical neural networks.

problem Time-consuming learning of MZI parameters in optical neural networks.
method Customized complex-valued derivatives and a chain rule for Wirtinger derivatives, incorporated into a function module.
result 20 times faster learning compared to conventional AD in MNIST task.

dYdX updates liquidity provider incentives to enhance trading efficiency.

problem Incentivizing liquidity providers to maintain efficient market structures.
method Analyzed various metrics (makerVolume, depths, spreads) and used historical trades to update the LP Incentives Programme.
result Updated the LP Incentives Programme to encourage more active and efficient liquidity.

Develops inference combinators for probabilistic programs using neural networks.

problem Creating efficient proposals for probabilistic program inference.
method Inference combinators using neural network parameterization of proposals.
result Correct by construction variational methods tailored to specific models.

A hardware-based reservoir computing system predicts time series with high speed and accuracy.

problem Processing time-dependent signals with high speed and accuracy.
method A hardware-based reservoir computing system using a field-programmable gate array (FPGA) for both the reservoir and output layers.
result Achieves comparable accuracy to software approaches but with a superior real-time prediction rate up to 160 MHz.

The paper gives a review of progress towards extending the Thurston programme to the Poincare duality case. For a full abstract, see the published version at the above link.

2004-10-03abs ↗pdf ↗

FixyNN splits CNN models into fixed and trainable parts for efficient on-device inference.

problem Energy inefficiency in on-device CNN inference for real-time computer vision.
method Co-designed hardware accelerator platform with transfer learning for training.
result Achieved nearly 2x better energy efficiency than a conventional accelerator.

This study categorizes RWA tokenization challenges and solutions.

problem Navigating the gap between on-chain deterministic code and off-chain probabilistic reality.
method Taxonomy and comparative analysis of RWA protocols, legal and technical standards.
result RWA tokenization requires overcoming legal and technical interoperability issues.

PDSketch enables flexible robot planning by learning from domain structures.

problem Building general robots with flexible planning.
method Exploiting locality and sparsity in environmental models, PDSketch defines high-level structures for trainable neural networks.
result PDSketch automatically generates planning heuristics without additional training.

Study optimizes CT and microinsurance for efficient social protection in low-income countries.

problem Efficient targeting of cash transfers to reduce social protection costs in low-income countries.
method Modelled household capital dynamics using piecewise-deterministic Markov process, derived HJB equation for optimal injection, used dynamic programming.
result Optimal level of capital injection above poverty threshold for cost-effective social protection.

New approach uses Boolean circuits to optimize neural networks.

problem Improving efficiency of neural network implementations on hardware accelerators.
method Formalized neural networks as Boolean circuits, showing binarized networks are functionally complete.
result Binarized neural networks are functionally complete, suggesting new possibilities for neural network accelerators.

In this study, after introducing algebraic properties of real quaternions some characterizations of quaternionic involute-evolute curves in Q are obtained. And some results and theorems for quaternionic w-curves are given. Lastly, we illustrate some examples and draw their figures with Mathematica Programme.

2013-11-04abs ↗pdf ↗

Transformers can emulate various algorithms by prompting, proving universality.

problem How to emulate algorithms using fixed-weight Transformers.
method Two modes of in-context algorithm emulation: task-specific and prompt-programmable. Constructing prompts that encode algorithm parameters into token representations.
result Fixed-weight Transformers can emulate a broad class of algorithms via prompts.

LeFlow enables quick FPGA synthesis from Tensorflow models.

problem Manual translation of Tensorflow models to FPGA RTL is time-consuming and requires expertise.
method Uses XLA compiler to emit synthesizable LLVM code from Tensorflow specifications, which is then synthesized.
result Allows users to generate Deep Neural Networks with just a few lines of Python code.

In these expository notes we draw together and develop the ideas behind some recent progress in two directions: the treatment of finite type partial differential operators by prolongation, and a class of differential complexes known as detour complexes. This elaborates on a lecture given at the IMA Summer Programme ``S…

2006-12-21abs ↗pdf ↗

This paper proposes a new way to quantize classical mechanical systems. Here we use ALAG - programme to construct moduli space of half weighted Bohr - Sommerfeld lagrangian cycles of fixed volume which is our quantum phase space. "Dynamical correspondence" principle makes possible to prove that this ALAG - quantization…

2001-06-01abs ↗pdf ↗