Requests for name changes in the electronic proceedings will be accepted with no questions asked. However name changes may cause bibliographic tracking issues. Authors are asked to consider this carefully and discuss it with their co-authors prior to requesting a name change in the electronic proceedings. Use the Report an Issue link to request a name change.

The use of deep learning techniques has achieved significant progress for program synthesis from input-output examples. However, when the program semantics become more complex, it still remains a challenge to synthesize programs that are consistent with the specification. In this work, we propose SED, a neural program generation framework that incorporates synthesis, execution, and debugging stages. Instead of purely relying on the neural program synthesizer to generate the final program, SED first produces initial programs using the neural program synthesizer component, then utilizes a neural program debugger to iteratively repair the generated programs. The integration of the debugger component enables SED to modify the programs based on the execution results and specification, which resembles the coding process of human programmers. On Karel, a challenging input-output program synthesis benchmark, SED reduces the error rate of the neural program synthesizer itself by a considerable margin, and outperforms the standard beam search for decoding.

The traditional reliability design methods are imperfect because the designed systems aim at fewer faults, but once a fault happens, the systems might hard fail. To solve this problem, we present a self-maintenance machine (SMM), one that can maintain its functions flexibly even though faults occur. To achieve the capabilities of diagnosing and repair planning, a model-based approach that uses qualitative physics was proposed. Regarding the repair-executing capability, a control-type repair strategy was followed. A prototype of the SMM was developed, and it succeeded in maintaining its functions if the structure did not change.

This article summarizes a workshop entitled "Detecting, Repairing, and Preventing Human-Machine Miscommunication," held on 4 August 1996 in Portland, Oregon. The author presents the significant issues raised during the four specific workshop sessions. Research related to achieving robust interaction is an important subarea in AI. Early work concerned the correction of spelling or grammatical errors in a user's utterance so that the system could more easily match them against a fixed linguistic model; work has also been done in the area of speech recognition, attempting to find the best fit of a sound signal to legal sequences of linguistic objects. All these approaches have assumed that the system's model is always correct.

With the exception of plan repair, important topics related to the use of plans (robust execution, reactivity, monitoring, evaluation) have received significantly less consideration. In realistic domains, however, plan generation is only a small component of the overall package. In particular, plans must be updated in response to new information and requirements in a timely fashion to ensure that they remain viable and relevant. Plan execution involves more than blind adherence to previously generated plans. Rather, run-time decisions are made to adapt, initiate, or abandon plans and activities in response to current considerations within the operating environment.

Humans learn from their mistakes. When things go badly, we notice that something is amiss, figure out what went wrong and why, and attempt to repair the problem. Artificial systems depend on their human designers to program in responses to every eventuality and therefore typically don't even notice when things go wrong, following their programming over the proverbial, and in some cases literal, cliff. This article describes our past and current work on the metacognitive loop, a domain-general approach to giving artificial systems the ability to notice, assess, and repair problems. The goal is to make artificial systems more robust and less dependent on their human designers.

There's a graveyard in space littered with the corpses of dozens of dead satellites, a remote spot in the cosmos reserved to entomb spacecraft at the end of their lives. Even the most robust and expensive satellites eventually break down or run out of fuel, and must be retired to a remote parking orbit more than 22,000 miles away, safely out of the way of other satellites. There, the graveyard holds billions of dollars-worth of some of the most expensive hardware ever to leave the surface of the Earth -- including not just commercial communications satellites, but some of the Pentagon's most sensitive assets, used for spying, guiding bombs and warning against missile launches. Now, the Defense Advanced Projects Agency, NASA and others, are developing technologies that would extend the life of the critical infrastructure in space, preventing satellites from being shipped to the graveyard for years. If successful, the agencies would have fleets of robots with arms and cameras that could inspect, refuel and repair satellites keeping them operational well beyond their expected lifetimes.

As a reasonably new person to the field of Machine Learning and programming in general (working in Python for 6 months now), I wanted to go to this forum and ask for some feedback and suggestions for an assignment I just started. For this assignment I have to look into machine-learning possibilities that automatically classify incoming invoices to decide what kind of repair was done on a car. There are between 1030-1500 different outcomes. Some occur very frequently, some almost never. The product code and product description probably contain the most useful information.

Using historic and live sensor data, the AI can do things like detect anomalies, predict when a machine will stop working, and help with supply chain management. Oracle is one of a range of companies -- including IBM and Microsoft -- that now offer industrial IoT solutions augmented by AI for training, safety, or machine maintenance. There's also a new Digital Fleet Management service for real-time shipment tracking and supply chain management and Smart Connected Factory, which performs root cause analysis and incident detection in industrial settings. "If a company is selling a pump or an HVAC or an elevator and they want to monitor those and offer a product as a service, the monitoring applications are focused on both the B2B and the B2C use cases for products," VP of industry and IoT solutions Lionel Chocron told VentureBeat.

When machine learning algorithms are used in life-critical or mission-critical applications (e.g., self driving cars, cyber security, surgical robotics), it is important to ensure that they provide some high-level correctness guarantees. We introduce a paradigm called Trusted Machine Learning (TML) with the goal of making learning techniques more trustworthy. We outline methods that show how symbolic analysis (specifi- cally parametric model checking) can be used to learn the dynamical model of a system where the learned model satis- fies correctness requirements specified in the form of temporal logic properties (e.g., safety, liveness). When a learned model does not satisfy the desired guarantees, we try two approaches: (1) Model Repair, wherein we modify a learned model directly, and (2) Data Repair, wherein we modify the data so that re-learning from the modified data will result in a trusted model. Model Repair tries to make the minimal changes to the trained model while satisfying the properties, whereas Data Repair tries to make the minimal changes to the dataset used to train the model for ensuring satisfaction of the properties. We show how the Model Repair and Data Repair problems can be solved for the case of probabilistic models, specifically Discrete-Time Markov Chains (DTMC) or Markov Decision Processes (MDP), when the desired properties are expressed in Probabilistic Computation Tree Logic (PCTL). Specifically, we outline how the parameter learning problem in the probabilistic Markov models under temporal logic constraints can be equivalently expressed as a non-linear optimization with non-linear rational constraints, by performing symbolic transformations using a parametric model checker. We illustrate the approach on two case studies: a controller for automobile lane changing, and query router for a wireless sensor network.